Electromagnetic detection optimized bipolar SHMPWM emission method

Abstract

The invention discloses an electromagnetic detection optimized bipolar SHMPWM emission method, which comprises the following steps: S1, by analyzing the influence of resistive and inductive loads on emission current waveform distortion and emission spectrum energy loss, extending a current waveform based on an equal-area principle, compensating lost emission current spectrum energy, and calculating current waveform continuation time; s2, considering an inductive load and a dead zone effect, and establishing a bipolar SHMPWM nonlinear equation set; and S3, performing iterative operation on the bipolar SHMPWM nonlinear equation set by using a particle swarm algorithm to obtain a switching angle solving result of the half-cycle symmetric SHMPWM. According to the method, an equation set is solved by adopting an optimized particle swarm algorithm to obtain a switching angle, so that the frequency domain information of a half-cycle symmetric SHMPWM frequency focusing waveform conforms to a specific harmonic amplitude expected to be set.

Description

Technical Field

[0001] The invention relates to the technical field of electromagnetic detection, and in particular to an electromagnetic detection optimized bipolar SHMPWM emission method. Background Art

[0002] The application of electromagnetic detection technology in mineral exploration, geological monitoring, engineering site selection, etc. is a research hotspot in the field of detection at home and abroad. Frequency domain electromagnetic detection method obtains geological information of specific strata by selecting appropriate transmission current frequency. Early studies proposed a multi-frequency method of simultaneously transmitting and measuring the fundamental frequency and multiple incidental spectra to improve detection efficiency. In order to further improve the controllability of the transmission waveform spectrum and reduce interference between adjacent frequencies, scholars have proposed methods such as m-sequence pseudo-random and 2n pseudo-random. These methods enhance frequency domain coverage by uniformly distributing the transmission waveform frequency. However, when conducting fine detection of specific target depths, the traditional multi-frequency transmission method still cannot meet the requirements and cannot obtain multiple closely spaced frequency signals, which limits the detection accuracy.

[0003] To address this problem, Selective Harmonic Elimination Pulse Width Modulation (SHMPWM) and Selective Harmonic Elimination Pulse Width Modulation (SHEPWM) are used as improved transmission strategies. By performing Fourier series expansion on the transmission waveform and solving a set of nonlinear equations to obtain the switching angle, the signal of a specific frequency is eliminated or retained. These methods can effectively eliminate adjacent harmonic interference, improve the accuracy and clarity of the spectrum, and are particularly suitable for fine detection of specific formations. Compared with conventional multi-frequency detection waveforms, SHMPWM and SHEPWM control methods have the following advantages: 1) Reduce switching frequency and switching loss; 2) Improve power quality and increase system efficiency; 3) Applicable to multi-level inverter design, expanding its application range; 4) Improve DC voltage utilization efficiency.

[0004] However, although the above methods have effectively improved the frequency controllability and detection accuracy of the transmission waveform, there are still some problems that have not been completely solved. For example, the transmission line inevitably has inductive reactance, which leads to the inconsistency of the speed of the current drop when the transmission current rises and turns off, resulting in the distortion of the transmission current waveform, which in turn causes the loss of spectrum energy. At the same time, the non-ideal change of the transmission current will produce strong primary field noise, which will interfere with the detection signal, especially the weak signals in the shallow and middle strata, which will also have an adverse effect on the time domain forward and inversion results of deep detection. In response to these problems, scholars have proposed optimization technologies such as segmented current control and PWM chopping, and designed improved transmission circuits such as absorption circuits, constant voltage clamping circuits and energy feedback circuits, and achieved certain results. However, due to the differences in the environments of different detection sites, the above optimization methods have insufficient adaptability, which always affects the accuracy and stability of electromagnetic detection. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide an electromagnetic detection optimized bipolar SHMPWM transmission method to achieve precise control of the output waveform DC and harmonic parameters while retaining only the required main frequency for transmission, thereby achieving efficient electromagnetic detection.

[0006] Technical solution: A method for optimizing bipolar SHMPWM transmission by electromagnetic detection, comprising the following steps:

[0007] S1, by analyzing the influence of resistive and inductive loads on the distortion of the emission current waveform and the loss of emission spectrum energy, based on the equal area principle, the current waveform is extended, the lost emission current spectrum energy is compensated, and the current waveform extension time is calculated;

[0008] S2, considering the inductive load and dead zone effect, establish the bipolar SHMPWM nonlinear equation group;

[0009] S3, using the particle swarm algorithm to iteratively calculate the bipolar SHMPWM nonlinear equation group, and obtain the switching angle solution of the half-cycle symmetrical SHMPWM.

[0010] Further, the specific steps for calculating the current waveform extension time are as follows:

[0011] S11, the transmitting circuit is equivalent to a first-order full-response resistive-inductive circuit;

[0012] From the first-order circuit full response circuit, we get:

[0013]

[0014] The topological relationship between the various variables and the resistive-inductive circuit is as follows:

[0015] f(t)=i(t)

[0016]

[0017] If the current is set to stabilize within 5τ, the relationship between t1 and t2 is as follows:

[0018] |t1-t2|>>5τ

[0019] Where f(t) is the instantaneous value of the first-order circuit current, f(∞) is the steady-state value of the first-order circuit current, and f(0 + ) is the initial value of the first-order circuit current, U d ,I d are steady-state voltage and steady-state current, L and R are the inductance and resistance of the equivalent circuit, t1 and t2 are the switch closing and opening times, i(t) is the on-time current, τ is the time constant, e is the natural constant, i(t 1+ ) is the initial value at the closing moment t1;

[0020] S12, based on the equal area principle, calculate the extension time ΔT of the unipolar actual current waveform;

[0021] The figure formed by the current waveform and the horizontal axis in the time period [t1, t1+5τ] is equivalent to a figure with a width of T1 and a height of I d rectangle, t1 is the starting time of the current rising phase, and T1 is solved using the following formula:

[0022]

[0023] Let t1 = 0, and we get:

[0024] T1=4τ+e -5 τ≈4τ

[0025] During the current drop period [t2, t3], the actual current i'(t) is:

[0026]

[0027] When the current drops to 0:

[0028]

[0029] Where, t2 is the starting time of the current drop phase, and t3 is the time when the current drops to 0;

[0030] The current waveform in the time period [t2, t3] and the horizontal axis are equivalent to a graph with a width of T2 and a height of I d Rectangle; from equal areas we get:

[0031]

[0032] Then we get T2 as:

[0033] T2=(1+ln(1 / 2))τ

[0034] According to the equal area principle, ΔT is widened on the basis of the original time from t1 to t2, so that the energy of the corrected emission current is equal to the energy of the ideal emission current; then ΔT is:

[0035] ΔT=5τ-T1-T2=ln2·τ

[0036] Step 13, calculating the extension time ΔT′ of the bipolar current waveform;

[0037] When the current is in the rising time period [t′1, t′1+5τ], the algebraic sum of the missing area and the excess area is equivalent to a width of ΔT′ and a height of I d The following formula is used to solve for ΔT′:

[0038]

[0039] Then we have:

[0040]

[0041] Let t′1=0, we get:

[0042] ΔT′=2ln2·τ

[0043] In the formula, t′1 and t′2 are the starting time and zero-crossing time of the rising phase of the actual bipolar current waveform respectively;

[0044] Applied to the equal area principle, the waveform from t′1 to t′3 should be widened by ΔT′ on the original basis, so that the corrected emission current energy is equal to the ideal emission current energy, and t′3 is the starting time of the descending phase of the actual bipolar current waveform.

[0045] Furthermore, the bipolar SHMPWM nonlinear equations are as follows:

[0046]

[0047] The relationship between the switch angle limit size is as follows:

[0048] 0<α1<α2<…<α N <π

[0049] Amplitude of each harmonic A i The calculation formula is as follows:

[0050]

[0051] Among them, a0 is the DC component, a i, b i are the sine and cosine components respectively, U SHM (ωt) is the instantaneous value of the inverter output voltage, U d is the inverter output DC voltage, α i is the switching angle, i=1,2,…,N, N is the number of switching angles in a half cycle; α d is the dead zone angle occupied by the dead zone time in a single cycle, T d is the dead time; ω is the angular velocity.

[0052] Further, in step S3, the optimized particle swarm algorithm is applied to solve the SHMPWM nonlinear equations, and the switching angles α1, α2, .., α N Expressed as follows:

[0053]

[0054] Fitness f i The definition is as follows:

[0055]

[0056] In the formula, ψ i () represents the switching angle α1, α2, .., α N The function of P N is a real constant obtained from the frequency domain characteristics of the given half-cycle mirror-symmetric SHMPWM waveform;

[0057] The calculation formula of adaptive weight g is as follows:

[0058]

[0059] In the formula, g max , g min are the maximum and minimum weight coefficients respectively; is the fitness of the i-th particle at the k-th iteration; is the average fitness of all particles at the kth iteration; is the minimum fitness of all particles at the kth iteration;

[0060] Taking into account the influence of adaptive weight g, acceleration factors C1 and C2 on the particle swarm algorithm, the optimized speed formula is as follows:

[0061]

[0062] In the formula, Rand() is a random number. represents the position of the i-th particle in the d-th dimension, is the optimal position of the current population, Best location for all populations;

[0063] X is the shrinkage factor, and the expression is as follows:

[0064]

[0065] Among them, C=C1+C2.

[0066] Compared with the prior art, the present invention has the following significant effects:

[0067] 1. Apply SHMPWM technology to electromagnetic detection, consider the influence of inductive load and dead zone effect of actual circuit on transmission waveform, and solve the problem that electromagnetic detection energy focusing and fine distribution cannot be effectively realized due to energy loss of transmission current spectrum; in view of the demand for fine exploration of underground mineral resources, consider the influence of inductive load and dead zone effect on harmonic distribution in actual SHMPWM transmission circuit, establish a half-cycle symmetrical SHMPWM nonlinear equation group, use optimized particle swarm algorithm to solve the equation group to obtain the switching angle, and realize that the frequency domain information of the half-cycle symmetrical SHMPWM frequency focusing waveform meets the expected specific harmonic amplitude;

[0068] 2. The present invention analyzes the influence of inductive load on the distortion of emission current waveform and spectrum energy loss, proposes a method of extending the current waveform based on the equal area principle, compensates for the lost emission current spectrum energy, and gives a method for calculating the extension time; at the same time, the dead zone effect is taken into account to achieve precise control of the DC and harmonic parameters of the output waveform, while retaining only the required main frequency for transmission, thereby achieving efficient electromagnetic detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a schematic diagram of the topological structure of the transmitting circuit that outputs a half-cycle symmetrical SHMPWM waveform;

[0070] Figure 2 It is a flow chart of the electromagnetic detection optimization bipolar SHMPWM transmission strategy considering the inductive load and dead zone effect;

[0071] Figure 3 (a) is the bipolar SHMPWM waveform without adding symmetrical dead zone.

[0072] (b) is the output half-cycle symmetrical SHMPWM waveform after adding symmetrical dead zone;

[0073] Figure 4 This is the ideal emission current waveform and actual emission current waveform of unipolar SHMPWM;

[0074] Figure 5 This is the ideal emission current waveform and actual emission current waveform of bipolar SHMPWM;

[0075] Figure 6 (a) is the waveform of the falling current of the resistive-inductive bipolar SHMPWM before optimization compensation, and (b) is the waveform of the falling current of the resistive-inductive bipolar SHMPWM after optimization compensation;

[0076] Figure 7 (a) is the rising current waveform of the resistive-inductive bipolar SHMPWM before optimization compensation, and (b) is the rising current waveform of the resistive-inductive bipolar SHMPWM after optimization compensation;

[0077] Figure 8 It is the flow chart of optimizing particle swarm algorithm solution;

[0078] Fig. 9 It is the frequency domain information diagram of the actual half-cycle symmetrical SHMPWM current before optimization;

[0079] Fig.10 It is the frequency domain information diagram of the optimized half-cycle symmetrical SHMPWM current. DETAILED DESCRIPTION

[0080] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0081] The present invention uses the specific harmonic weakening pulse width modulation (SHMPWM) technology to establish a mathematical model and designs a method for outputting an optimal emission current to achieve electromagnetic detection energy focusing and fine distribution.

[0082] like Figure 1 As shown in the figure, a diode clamped three-level inverter is used as the transmitter circuit topology. The three-level inverter has less switch tube stress and lower loss than the two-level inverter, which can meet the needs of high-power and high-voltage output of electromagnetic detection transmitters. The inverter contains 8 power switch tubes S1-S8, 8 reverse parallel freewheeling diodes D1-D8, and four clamping diodes V D1 -V D8 And two clamping capacitors C1 and C2. C1 and C2 use the same specification capacitors, so that N o The load end of the electromagnetic transmitter is connected to the Figure 1 Between points A and B. e is the supply voltage, i o 、u o The transmitter outputs a rectangular voltage u o Through the cable and grounding electrodes a and b, an emission current i is generated in the earth load. o At the same time, due to the rapid change of the polarity of the transmitting current, induction is generated deep in the earth, and the induced signal is radiated to the ground, which is received and analyzed by the receiver to obtain the formation information.

[0083] like Figure 2 As shown in FIG. 1 , it is a flow chart of a bipolar SHMPWM transmission method for electromagnetic detection optimization considering the inductive load and dead zone effect. The SHMPWM waveform is modulated offline through the pre-stored switching angle, taking the dead zone problem into consideration. Figure 3 As shown in (a) and (b) in FIG. 1 , the output half-cycle symmetrical SHMPWM waveform is shown before and after the symmetrical dead zone is added.

[0084] The specific steps of the electromagnetic detection optimization bipolar SHMPWM transmission method considering the inductive load and dead zone effect of the present invention are as follows:

[0085] Step 1, based on the equal area principle, extend the current waveform and calculate the current waveform extension time;

[0086] By analyzing the influence of resistive and inductive loads on the distortion of the emission current waveform and the loss of emission spectrum energy, the current waveform is extended based on the equal area principle, and the emission current spectrum energy method that compensates for the loss is used to calculate the current waveform extension time.

[0087] The specific steps are as follows:

[0088] Step 11, the transmitting circuit is equivalent to a first-order full-response resistive-inductive circuit;

[0089] From the full response of the first-order circuit, we can get:

[0090]

[0091] The topological relationship between the various variables and the resistive-inductive circuit is as follows:

[0092] f(t)=i(t) (2)

[0093]

[0094] The current when it is turned on is:

[0095]

[0096] If the current is set to stabilize within 5τ, the relationship between t1 and t2 is as follows:

[0097] |t1-t2|>>5τ (6)

[0098] Where f(t) is the instantaneous value of the first-order circuit current, f(∞) is the steady-state value of the first-order circuit current, and f(0 + ) is the initial value of the first-order circuit current, U d ,I dare steady-state voltage and steady-state current, L and R are the inductance and resistance of the equivalent circuit, t1 and t2 are the switch closing and opening times, i(t) is the on-time current, τ is the time constant, e is the natural constant, i(t 1+ ) is the initial value at the moment of closing at time t1.

[0099] Step 12, based on the equal area principle, calculating the extension time ΔT of the unipolar actual current waveform;

[0100] like Figure 4 The figure shows the unipolar ideal emission current waveform and the actual emission current waveform. The current waveform in the time period [t1, t1+5τ] and the horizontal axis are now equivalent to a graph with a width of T1 and a height of I d rectangle, t1 is the starting time of the current rising phase, and T1 is solved using the following formula:

[0101]

[0102] Let t1 = 0, and we get:

[0103] T1=4τ+e -5 τ≈4τ (8)

[0104] During the current drop period [t2, t3], the actual current i'(t) of the transmitting circuit is:

[0105]

[0106] When the current drops to 0:

[0107]

[0108] Wherein, t2 is the starting time of the current decrease phase, and t3 is the time when the current drops to 0.

[0109] Similarly, the current waveform of the time period [t2, t3] and the horizontal axis are equivalent to a graph with a width of T2 and a height of I d Rectangle. Since the areas are equal, we can get the following formula:

[0110]

[0111] Then we get T2 as:

[0112] T2=(1+ln(1 / 2))τ (12)

[0113] Applying the equal area principle, the original time from t1 to t2 is widened by ΔT, so that the energy of the corrected emission current is equal to the energy of the ideal emission current. Then ΔT is:

[0114] ΔT=5τ-T1-T2=ln2·τ (13)

[0115] Step 13, calculating the extension time ΔT′ of the bipolar current waveform by analogy with the calculation method of the extension time of the unipolar current waveform;

[0116] like Figure 5 is the bipolar ideal emission current waveform and the actual emission current waveform. When the current is within the rise time [t′1, t′1+5τ], the algebraic sum of the missing area and the excess area is equivalent to a width of ΔT′ and a height of I d The following formula is used to solve for ΔT′:

[0117]

[0118] Similar to the unipolar resistive inductive load waveform, there are:

[0119]

[0120] Let t′1=0, we get:

[0121] ΔT′=2ln2·τ (16)

[0122] In the formula, t′1 and t′2 are the starting time and zero-crossing time of the rising phase of the actual bipolar current waveform respectively;

[0123] Applied to the equal area principle, the waveform from t′1 to t′3 should be widened by ΔT′ on the basis of the original waveform, so that the energy of the corrected emission current is equal to the energy of the ideal emission current. t′3 is the starting time of the descending phase of the actual bipolar current waveform. The emission current waveforms before and after correction are as follows Figure 6 As shown in (a) and (b), the black shaded area is the part of the actual current that needs to be compensated.

[0124] Similarly, the compensation waveform of the bipolar current in the rising section can be obtained. The compensation time is ΔT′. The rising current waveform of the resistive-inductive bipolar SHMPWM before and after compensation is as follows: Figure 7 As shown in (a) and (b) in .

[0125] Step 2, taking the dead zone effect into account, establishing an optimized bipolar SHMPWM nonlinear equation group considering the inductive load and the dead zone effect;

[0126] The nonlinear equations are as follows:

[0127]

[0128] The switch angle limit size relationship is as follows:

[0129] 0<α1<α2<…<α N <π (18)

[0130] α d With T d The relationship is as follows:

[0131]

[0132] Amplitude of each harmonic A i The calculation formula is as follows:

[0133]

[0134] Among them, a0 is the DC component, a i , b i are the sine and cosine components respectively, U SHM (ωt) is the instantaneous value of the inverter output voltage, U d is the inverter output DC voltage, α i is the switching angle, i=1,2,…,N, N is the number of switching angles in a half cycle, α d is the dead zone angle occupied by the dead zone time in a single cycle, T d is the dead time, and ω is the angular velocity.

[0135] Step 3, using the optimized particle swarm algorithm to solve the optimal solution of the nonlinear equations, so as to obtain the SHMPWM specific switching angle that meets the actual requirements;

[0136] By optimizing the particle swarm algorithm, a strategy is designed that can automatically exit the iterative cycle when the fitness meets the requirements. The flow chart is as follows Figure 8 As shown, the specific steps are as follows:

[0137] Step 31, initialize the maximum number of iterations Max, the count setting value count_set and the counter counter; set the function variation tolerance F_tolerance, which is generally a very small positive number;

[0138] Step 32, in each iteration, the best fitness f_best is calculated, and the absolute value f_change of the change between the best fitness of the current iteration and the best fitness of the previous iteration is calculated;

[0139] Step 33, determine whether the absolute value of the fitness change f_change is less than the function change tolerance F_tolerance, if so, the counter counter is incremented by 1, otherwise the counter counter is reset to zero;

[0140] Step 34, if the current number of iterations exceeds the maximum number of iterations Max, then the loop is exited to obtain the optimal solution; if the current number of iterations does not exceed the maximum number of iterations Max, and the count value is greater than the count setting value count_set, it means that the optimal solution is within a certain fluctuation range, then the loop is directly exited;

[0141] Step 35, determine whether the best fitness f_best meets the solution accuracy. If not, continue to loop steps 31 to 34 until the accuracy is met and the optimal solution is obtained, and then end the operation.

[0142] The optimized particle swarm algorithm is used to solve the SHMPWM nonlinear equations and the switching angles α1, α2, .., α N Expressed as follows:

[0143]

[0144] Fitness f i The definition is as follows:

[0145]

[0146] In the formula, ψ i represents the switching angles α1, α2, .., α N The function of P N It is a real constant obtained from the frequency domain characteristics of the given half-cycle mirror-symmetric SHMPWM waveform.

[0147] In terms of computing performance, an adaptive weight is used in combination with a shrinkage factor. For problems that require rapid convergence to a local optimal solution, the adaptive weight can be reduced; for problems that require extensive search for a global optimal solution, the adaptive weight should be increased. The present invention dynamically adjusts the adaptive weight according to the search process to balance the global and local search capabilities. The adaptive weight g is calculated as follows:

[0148]

[0149] In the formula, g max , g min They are the maximum and minimum weight coefficients, which are usually 0.9 and 0.4; is the fitness of the i-th particle at the k-th iteration; That is, the average fitness of all particles at the kth iteration;

[0150] That is, the minimum fitness of all particles at the kth iteration;

[0151] Taking into account the influence of adaptive weight g, acceleration factors C1 and C2 on the particle swarm algorithm, the optimized speed formula is as follows:

[0152]

[0153] In the formula, Rand() is a random number. represents the position of the i-th particle in the d-th dimension, is the optimal position of the current population (referring to the position of the individual with the best performance among all individuals in the current population), is the best position of all populations (referring to the position of the optimal solution found in the population so far in the entire optimization process).

[0154] A new influencing factor, the contraction factor x, is introduced into the new speed formula as shown below:

[0155]

[0156] Among them, C=C1+C2.

[0157] In practice, it is found that within a certain range, the higher the speed of the particles themselves, the stronger the performance of the particle swarm algorithm and the better the effect. The contraction factor x has a similar effect to the adaptive weight g. The choice of which parameter to use should depend on the problem to be solved. The speed of the particles can be controlled by setting the contraction factor x and the adaptive weight g parameters, thereby improving the effect of the particle swarm algorithm. The acceleration factors C1 and C2 are usually set to 2. In fact, in order to better control the convergence speed and avoid premature convergence, the value of C is usually selected to be slightly larger than 4, so that a more appropriate contraction factor value can be obtained. This particle swarm algorithm takes C≈4.1, at which time x≈0.729.

[0158] In order to avoid falling into the local optimum, the tolerance of the function change is set as a related quantity. Before each loop, the absolute value of the fitness change is calculated for comparison. If the absolute value of the fitness change is always less than the preset tolerance, it means that the search particle is in the best position, the solution tends to converge, and the loop can be automatically jumped out. In terms of solving the angle constraint, considering the actual situation, the emission setting of the SHMPWM waveform is carried out, and the constraint conditions for solving the switch angle are added. If the requirements are not met, the particle swarm is initialized repeatedly for recalculation. In terms of solving the accuracy, the accuracy requirements are set. The optimal solution obtained by the optimization still does not meet the high-precision setting value, and the operation has to be repeated until the solution value obtains a satisfactory effect. Figure 8 Flowchart for optimizing the particle swarm algorithm solution.

[0159] Based on the same inventive concept, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the processor implements the aforementioned steps of optimizing the bipolar SHMPWM emission strategy by electromagnetic detection taking into account the inductive load and dead zone effects.

[0160] Table 1 shows the switching angle solution results of the half-cycle symmetrical SHMPWM considering the inductive load and dead zone effect.

[0161] Table 1. Solution results of half-cycle symmetrical SHMPWM switching angle

[0162] Switching angle <![CDATA[ α1 ]]> <![CDATA[ α2 ]]> <![CDATA[ α3 ]]> <![CDATA[ α4 ]]> <![CDATA[ α5 ]]> <![CDATA[ α6 ]]> <![CDATA[ α7 ]]> <![CDATA[ α8 ]]> <![CDATA[ α9 ]]> <![CDATA[ α10 ]]> After optimization 0.4388 0.6155 0.8361 1.0596 1.4978 1.6438 2.0820 2.3055 2.5261 2.7028

[0163] Fig. 9 This is the actual half-cycle symmetrical SHMPWM current frequency domain information diagram before optimization. Fig.10 This is the half-cycle symmetrical SHMPWM current frequency domain information diagram after optimization considering the actual current waveform extension time and dead zone effect. Set the output base frequency to 50Hz, the current harmonic amplitude to 0.8A, and retain the 1st, 5th, and 7th harmonics. Fig. 9 , Fig.10 Comparative analysis shows that the amplitudes of the 5th and 7th harmonics are obviously compensated after optimization, which is basically consistent with the ideal situation.

[0164] The above embodiments are only for illustrating the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for optimizing bipolar SHMPWM transmission by electromagnetic detection, characterized in that: The steps include: S1, by analyzing the influence of resistive and inductive loads on the distortion of the emission current waveform and the loss of emission spectrum energy, based on the equal area principle, the current waveform is extended, the lost emission current spectrum energy is compensated, and the current waveform extension time is calculated; S2, considering the inductive load and dead zone effect, establish the bipolar SHMPWM nonlinear equation group; S3, using the particle swarm algorithm to iteratively calculate the bipolar SHMPWM nonlinear equation group, and obtain the switching angle solution of the half-cycle symmetrical SHMPWM.

2. The electromagnetic detection optimization bipolar SHMPWM transmission method according to claim 1 is characterized in that: The specific steps for calculating the extension time of the current waveform are as follows: S11, the transmitting circuit is equivalent to a first-order full-response resistive-inductive circuit; From the first-order circuit full response circuit, we get: The topological relationship between the various variables and the resistive-inductive circuit is as follows: f(t)=i(t) If the current is set to stabilize within 5τ, the relationship between t1 and t2 is as follows: |t1-t2|>>5τ Where f(t) is the instantaneous value of the first-order circuit current, f(∞) is the steady-state value of the first-order circuit current, and f(0 + ) is the initial value of the first-order circuit current, U d ,I d are steady-state voltage and steady-state current, L and R are the inductance and resistance of the equivalent circuit, t1 and t2 are the switch closing and opening times, i(t) is the on-time current, τ is the time constant, e is the natural constant, i(t 1+ ) is the initial value at the closing moment t1; S12, based on the equal area principle, calculate the extension time ΔT of the unipolar actual current waveform; The figure formed by the current waveform and the horizontal axis in the time period [t1, t1+5τ] is equivalent to a figure with a width of T1 and a height of I d rectangle, t1 is the starting time of the current rising phase, and T1 is solved using the following formula: Let t1 = 0, and we get: T1=4τ+e -5 τ≈4τ During the current drop period [t2, t3], the actual current i'(t) is: When the current drops to 0: Where, t2 is the starting time of the current drop phase, and t3 is the time when the current drops to 0; The current waveform in the time period [t2, t3] and the horizontal axis are equivalent to a graph with a width of T2 and a height of I d Rectangle; from equal areas we get: Then we get T2 as: T2=(1+ln(1 / 2))τ According to the equal area principle, ΔT is widened on the basis of the original time from t1 to t2, so that the energy of the corrected emission current is equal to the energy of the ideal emission current; then ΔT is: ΔT=5τ-T1-T2=ln2·τ Step 13, calculating the extension time ΔT′ of the bipolar current waveform; When the current is in the rising time period [t′1, t′1+5τ], the algebraic sum of the missing area and the excess area is equivalent to a width of ΔT′ and a height of I d The following formula is used to solve for ΔT′: Then we have: Let t′1=0, we get: ΔT′=2ln2·τ In the formula, t′1 and t′2 are the starting time and zero-crossing time of the rising phase of the actual bipolar current waveform respectively; Applied to the equal area principle, the waveform from t′1 to t′3 should be widened by ΔT′ on the original basis, so that the corrected emission current energy is equal to the ideal emission current energy, and t′3 is the starting time of the descending phase of the actual bipolar current waveform.

3. The electromagnetic detection optimization bipolar SHMPWM transmission method according to claim 2 is characterized in that: The bipolar SHMPWM nonlinear equations are as follows: The relationship between the switch angle limit size is as follows: 0＜α1＜α2＜…＜α N ＜π Amplitude of each harmonic A i The calculation formula is as follows: Among them, a0 is the DC component, a i 、b i are the sine and cosine components respectively, U SHM (ωt) is the instantaneous value of the inverter output voltage, U d is the inverter output DC voltage, α i is the switching angle, i=1,2,…,N, N is the number of switching angles in a half cycle; α d is the dead zone angle occupied by the dead zone time in a single cycle, T d is the dead time; ω is the angular velocity.

4. The electromagnetic detection optimization bipolar SHMPWM transmission method according to claim 3 is characterized in that: In step S3, the optimized particle swarm algorithm is used to solve the SHMPWM nonlinear equations and the switching angles α1, α2, .., α N Expressed as follows: Fitness f i The definition is as follows: In the formula, ψ i () represents the switching angle α1, α2, .., α N The function of P N is a real constant obtained from the frequency domain characteristics of the given half-cycle mirror-symmetric SHMPWM waveform; The calculation formula of adaptive weight g is as follows: In the formula, g max , g min are the maximum and minimum weight coefficients respectively; is the fitness of the i-th particle at the k-th iteration; is the average fitness of all particles at the kth iteration; is the minimum fitness of all particles at the kth iteration; Taking into account the influence of adaptive weight g, acceleration factors C1 and C2 on the particle swarm algorithm, the optimized speed formula is as follows: In the formula, Rand() is a random number. represents the position of the i-th particle in the d-th dimension, is the optimal position of the current population, Best location for all populations; X is the shrinkage factor, and the expression is as follows: Among them, C=C1+C2.

Images