Development of a 300W Generator for Lightweight Wind Turbine

Ⅰ. Introduction

Population growth in leisure activities has brought about a good deal of demand for portable and eco-friendly power generation systems. Small wind power systems have emerged as one of the alternatives to fill the demands[1].

Research on the existing permanent magnet type generators has been conducted mainly to increase the efficiency and reduce the cogging torque. In this paper, a method of reducing the weight of a generator at the same efficiency was studied.

The most important factor in developing a small wind generator is to increase the efficiency by reducing the weight while augmenting ease of movement. The existing 300W wind power generators are quite heavy to carry, weighing at about 10kg. The weight of these wind power generators can be reduced to less than 4kg for portability. It can also be developed in a smaller form that can be used in any place regardless of wind strengths. Considering the various types of air volume and geographical features of Korea, a smaller and lighter wind power generator that can be used independently has been developed.

Ⅱ. Design of 300W Wind Turbine Generator System

Since a surface permanent magnet synchronous motor (hereinafter SPMSM) does not require field strengths and field currents, it is possible to make the structure of a wind turbine generator simpler and smaller which leads to decrement of energy losses while increasing operational efficiency. It has the advantage of decreasing cogging torque and torque ripple effect; both of which affect starting torque due to the low effect from harmonics. In light of the benefits of SPMSM, a design of it was developed in this research[2]-[4].

The final, lighter 300W wind turbine generator system was planned by determining a fundamental structure of the generator through basic designing process and decided the detailed size of the permanent magnet[5].

2.1 Basic Design of Wind Turbine Generator System

The electrical loading method was conducted for the 300W wind turbine generator as the basic designing[4]. In this process, combinations of poles and slots, stator wiring, and shapes of stators and rotators were planned.

SPMSM designing requires a relevant selection of polar and slots in accordance with the capacity and purpose of the device. Normally, the number of slots is determined by proportional numbers to the poles and the number of poles is determined by the proportional even number.

Among the combinations of poles and slots, there are ones that are frequently used depending on the amount of cogging torque and outputs. In the case of a ratio of three to three, a smaller generator typically uses eight to twelve poles. In this work, a combination of eight poles and eighteen slots is decided in consideration of the least cogging torque that can generate electricity in the light wind. The ratio of the slots and poles is 2.25[6][7].

With the combination of poles and slots, a total of eighteen coils are designed and six coils are assigned to each phase. These six coils are separated in two of three coils and each phase consists of two coil bundles that are one in a finely spaced coil arrangement. The bundles of coils are winded in different directions as shown in Fig. 1 the Coil 1 and 3 are winded in the same direction whereas the Coil 2 is winded in the opposite.

Fig. 1.

Coil of one phase coil

The initial shape of the generator is shown in Fig. 2. Table 1 shows output power and efficiency in accordance with sizes of frame, stack thickness, and weights of the generator. The sizes of frames vary from 80m to 160m and thickness of stacks differs from 24m and 48m.

Fig. 2.

Initial shape of the generator

Table 1.

Output power and efficiency

The approximate sizes of stators, rotors, slots, and magnets are decided based on researchers experiences and resources. The figures of output power and efficiency came out via a calculation program. After checking the output power and efficiency of each trial, the basic shape of the generator was determined.

In planning the basic form of the generator, an approximate setting of a stator and a rotor was done at first. While the external diameter and the thickness were being fixed, detailed designing for the stator and the rotor was done.

Holding the outer diameter and maintaining the thickness with the initially determined figures, a detailed designing of the stator and the rotor was carried out.

Table 2 shows the specification of the generator design and Table 3 shows the results calculated with the design specification of Table 2.

Table 2.

Generator design specification

Table 3.

Calculation result

The figure of output power and efficiency in the case of the outer diameter at 140mm and stack thickness of 28mm appeared as the best way of meeting the goals of lighter weight and efficiency. Assuming this particular set of design would work with the detailed model later, it was determined as the foundation.

2.2 Optimum Design of Wind Turbine Generator System

In this process, a specific model for the ultimate wind generator was produced by conducting optimized designing based on the previous basic set. The purpose of this process was to reduce the negative influences of cogging and ripples of torque that affected the starting torque. There are various ways to prevent it. A stator or a rotor can be skewed, the structure of slots can be changed, the slot form can be changed, and auxiliary slots can be added. A study on a designing technique that reduces the influences of cogging torque and ripple effects by changing the shape of a permanent magnet was performed. A new technique based on the study, which uses the finite element analysis program and evolutionary optimization algorithm was developed. Using this technique, a shape of permanent magnet with the least influence of cogging and ripples of torque under 300W output power was produced[8]-[10].

Fig. 3 shows the optimized shape of the permanent magnet. The distribution of the optimization model computed by the finite element analysis (FEA) program is shown in Fig. 4. In the magnetic field distribution, there is about 1.2T in the gap between the stator and the rotor, about 1.7T in the middle of the tooth, and approximately 2T at the saturation point of the teeth are distributed.

Fig. 3.

Optimized shape of permanent magnet

Fig. 4.

Magnetic field distribution of the wind turbine optimization model

In the optimized model, the no-load maximum voltage of 35.5V indicates cogging torque over time. While the initial model of cogging torque is 68.4mNm, the optimization model’s is 1.9mNm with which the latter shows significant amount of decrease of cogging torque in comparison to the initial model. The total amount of torque ripples produced over time under the normal condition is 9.15% and 3.19% respectively.

Ⅲ. Manufacturing a 300W of wind turbines

Based on the design previously planned, a 300W scale wind turbine was produced. The manufacturing process of the generator was conducted in the order of laminating layers of iron core and welding, fixing the rotor magnets, winding the wires of the stator, and binding the rotor and the stator.

The material of the iron core used for the stator and the rotor is 35PN440. The iron core was made of several layers of iron plates laminated thicker than the original plan. Surface treatment and welding were done. After welding, the iron cores of side ends were removed and the thickness of the iron core was adjusted to match the thickness in accordance with the plan. Fig. 5 shows the iron cores of the stator one on the left, and the rotor one on the right.

Fig. 5.

Wind power generator stator iron core and rotor iron core

The magnets for the wind turbine are neodymium magnets N39SH with about 1.26T. The wind turbine was designed to have eight poles, four poles alternately being fixed on N pole and S pole. A thermal adhesive with epoxy resin was used to attach the magnets on the rotor and a heat treatment followed. Fig. 6 shows the procedure of the rotor magnet attachment from the left to the right.

Fig. 6.

Rotor magnet attachment process

The stator winding has 18 slots with 18 coils winding up each of the six coils as previously mentioned in the design section. Fig. 7 shows a picture of the wound stator. One side of the wound bundles of wires are connected to the other bundle in the shape of Y connection with one strand of wire per phase pulled out.

Fig. 7.

Coil-wound stator

Ⅳ. Characteristic Test of the Generator

Based on the basic designing and optimization process of the 300W scale wind turbine, the generator was produced. We carried out a characteristic test of the 300W wind turbine to check its cogging torque, output power of the motor, and the product specifications[11][12].

The basic specifications of wind power generators, resistance, inductance, and weight were measured. Fig. 8 shows the picture of the test. Picture (a) displays inductance measurement, (b) shows resistance measurement, and (c) shows the weight of measurement test. Measurements resulted in inductions at 4.7mH, line resistance 1.084Ω, and weight at 3.84kg.

Fig. 8.

300W class wind turbine generator specification measurement, (a) Inductance measurement test, (b) Resistance measurement test, (c) Weighting test

Cogging torque of the generator was measured. After the 300W scale wind generator test sample was tightened to the torque meter, it was measured from 0˚ to 360° clockwise while rotating it at 1rpm.

Fig. 9 shows the wind generator and cogging torque meter connected to the test measurement and Fig. 10 shows a graph of the torque output. Measured value of cogging torque was a minimum of -33mNm and a maximum of 69.2mNm with the average value of 51.3mNm. The waveforms of periodic torque generated by the windings and magnets were verified.

Fig. 9.

Cogging torque test

Fig. 10.

Graph of cogging torque measured

A no-load test and a load test were done. The load test was carried out at the rated load. The no-load test was to measure the counter electro-motive force when the generator is at no-load. After the sample was locked in the environmental chamber dynamo, the counter electromotive force was measured by a power analyzer while increasing the rotations per minute by 50 from 0 to 3,000.

Fig. 11 shows the sample for the no-load test after locking and Fig. 12 shows the measured data of the counter electromotive force in the no-load test. The rated operational standard of the designed wind turbine was set to 900rpm when the ambient wind direction was 12m/s. A 300W level wind turbine generator at a rated speed of 900rpm, the counter electromotive force of U was measured as 24.7V_rms, 24.9V_rms for V, and 24.9V_rms for W, and the average voltage was 24.9V_rms. The output voltage was 4.15V_rms at the minimum wind speed of the generator 150rpm and was 54.8V_rms, at the maximum wind speed 1998rpm.

Fig. 11.

Generator no-load test

Fig. 12.

Data of the No-Load counter EMF measurement

Fig. 13 shows the no-load output voltage of each phase measured by an oscilloscope, (a) with 150rpm, (b) with 900rpm, and (c) with 1998rpm. The no-load average output voltage of a 300W class wind turbine was about 4.15V_rms, which indicates that the output voltage of sine wave was inducted at the start wind speed. The average output voltage at the rated speed of 900rpm was about 24.7V_rms and 54.8V_rms was measured at the maximum wind speed of 1998rpm.

Fig. 13.

No-Load output voltage of each phase of 300W class wind turbine generator

The load test is to measure output voltage, current, and efficiency of the generator by measuring the electric power generated when the output terminals of U, V, and W of the sample are connected to the resistant delta wire and are locked to the environmental chamber dynamo rotating at 900rpm. Fig. 14 shows the load test of the generator. The test result shows that the average line voltage is 36.8V_rms and the average phase current is 2.62A. The power output was 289.5W when mechanical input was 339.84W and the efficiency of the generator is 85.18%.

Fig. 14.

Generator load test

Ⅴ. Conclusion

A characteristic test was conducted after manufacturing a lightweight wind turbine generator based on the design of wind turbines. The electric loading planning was used to construct the basic design of the generators such as the shape and size at the beginning stage of designing the wind generator and the optimum designing for the permanent magnet and torque was done at the optimization designing stage. In the latter stage, there was no change made on the basic shape of the generator, but the cogging torque was reduced from 68.4mNm to 1.9mNm and torque ripple was decreased from 9.15% to 3.19%.

Based on the results of the designing process, a wind turbine generator was manufactured and characteristic tests for the generator were carried out. The result of the no-load test for the 300W wind turbine generator weighing 3.84kg shows that its average output voltage is 24.7V_rms at the rated speed of 900rpm and its generating efficiency is more than 85%. This study enabled the design of wind turbines to be manufactured to produce more efficient power generation than conventional generators.

Acknowledgments

This work was supported by the Soonchunhyang University Research Fund.

References

• Jan olof Helldin, Jens Jung, Wiebke Neumann, Mattias Olsson, Annas Karin, and Fredrik Widemo, "The Impacts of wind power on terrestrial mammals", Naturvårdsverket report no 6499, Aug), (2012.
• C. C. Chan, and K. T. Chau, "Novel Permanent Magnet Motor Drives for Electric Vehicles", IEEE Trans. on Industrial Electronics, 43(2), p331-339, Apr), (1996. [https://doi.org/10.1109/41.491357]
• J. H. Kim, S. G. Kang, and S. Y. Jung, "Design of SPMSM for Eelectric Power Steering in vehicle Considering Cogging Torque reduction", 2015 KIEE 46th Conference, p123-124, May), (2015.
• T. Takeuchi, "DAIGAKUKATEI DENKI SEKKEIGAKU", Korean language edition, p27-64, Dong-Il publishing Co., (1994).
• J. G. Park, G. S. Cha, H. J. Lee, and Y. S. Kim, "Design of a wind turbine generator with low cogging torque by using evolution strategy", Journal of the Korean Academia-Industrial cooperation Society, 17(11), p755-760, Nov), (2016. [https://doi.org/10.5762/kais.2016.17.11.755]
• J. R. Hendershot Jr, and T. J. E. Miller, "Design of Brushless Permanent- Magnet Motors", p3-1-3-74, MagnaPhysics, Mar), (1994.
• H. J. Lee, Y. M. Kim, J. S. Oh, M. H. Ku, K. H. Baek, and G. S. Cha, "1kW Class of Small Wind Generators Cogging Torque Reduction Method of Research", 2012 KIEE Conference, p288-290, Nov), (2012.
• J. Kwak, S. Min, and J. P. Hong, (2010, "Optimal Stator Design of Interior permanent Magnet Motor to Reduce Torque Tipple Using Level Set Method", IEEE Trans. Magn., 46(6), p2108-2111, Jun), (2010.
• K. Yoo-Gyu, and K. Won, "Ubiquitous Sensor Network Design Using Genetic Algorithm Based on Ranking Selection Technique", The proceedings of the 2010 KIIT Summer Conference, p187-190, May), (2010.
• M. Kasper, "Shape Optimization by Evolution Strategy", IEEE Trans. on Magnetics, 28(2), p1556-1559, Feb), (1992. [https://doi.org/10.1109/20.123995]
• S. H. Thomas, T. M. Jahns, and W. I. Soong, "Torque Ripple Reduction in Interior Permanent Magnet Synchronous Machines Using the Principle of Mutual Harmonics Exclusion", Industry Applications Conference, p558-565, Sep), (2007. [https://doi.org/10.1109/07ias.2007.90]
• H. G. Kim, and J. S. Kong, "Characteristics Analysis of Permanent-Magnet Generator for Small Wind Turbine", 2009 KIEE 39th Conference, p708-709, Sep), (2009.