Traveling wave electric curtain adaptive control method based on SBM dust prediction

Abstract

The invention relates to the field of solar cell panel dust removal, and discloses a traveling wave electric curtain self-adaptive control method based on SBM dust prediction, which comprises the following steps: collecting photovoltaic cell panel direct-current power generation power, photovoltaic cell panel slope total irradiance, environment temperature and environment relative humidity during historical normal operation of traveling wave electric curtain dust removal equipment to form a standard parameter set; a state matrix D based on SBM is constructed, the parameter state of the traveling wave electric curtain in actual operation is obtained, and the parameter state of the traveling wave electric curtain in normal operation under the current working condition is predicted based on the actual operation parameter state and the state matrix D; and according to the actual running parameter state of the traveling wave electric curtain and the predicted parameter state during normal running, determining an abnormality factor of the parameter, judging the running state of the equipment according to the abnormality factor, and adaptively controlling the traveling wave electric curtain dust removal equipment. The voltage of the alternating current power supply can be adaptively controlled, so that the working energy consumption is reduced, and the application and popularization value of the electric curtain is improved.

Description

Technical Field

[0001] The present invention relates to the field of dust removal for solar panels, and mainly relates to a self-adaptive control method for traveling-wave electric curtains based on SBM dust prediction. Background Art

[0002] During the operation of a photovoltaic power station, solar panels are vulnerable to natural factors such as sand and dust, dust deposition, fouling, rain and fog. The problem of dust accumulation on the surface is serious, which may lead to a 20% - 40% decrease in power generation efficiency. The electric curtain dust removal technology has strong application potential and a better application prospect compared with other dust removal methods because it does not require other mechanical devices and manual operations. The working principle of this technology is to use high-voltage alternating current to drive dust particles to bounce under the action of the alternating electric field of the electric curtain, break away from the surface of the solar panel, and further move in a directional manner under the action of the electric field force, and finally leave the surface of the solar panel.

[0003] At present, the excitation of the electric curtain is very blind. The design of the electric curtain parameters mainly relies on the experimental "trial and error" method, and there is a lack of research on the design and control strategy of electric curtain parameters based on dust accumulation prediction and aiming at improving efficiency and reducing energy consumption. Summary of the Invention

[0004] Object of the Invention: To solve the problems existing in the prior art, the present invention proposes a self-adaptive control method for traveling-wave electric curtains based on SBM dust prediction. By using SBM modeling technology, according to the representative working state of the equipment when it is clean and dust-free, the state that the equipment should present during normal operation is estimated, and the degree of its abnormal operation is diagnosed. This abnormal degree represents the degree of dust accumulation, and then the voltage of the AC power supply is adaptively controlled to reduce its working energy consumption and improve the application and popularization value of the electric curtain.

[0005] Technical Solution: The present invention provides a self-adaptive control method for traveling-wave electric curtains based on SBM dust prediction, including the following steps:

[0006] Step 1: Collect the DC power generation power of the photovoltaic panel, the total irradiance on the inclined surface of the photovoltaic panel, the ambient temperature, and the ambient relative humidity when the equipment operates normally in history, and form a standard parameter set X. The normal operation of the equipment refers to the operation process of the traveling-wave electric curtain without dust accumulation.

[0007] Step 2: Based on the standard parameter set X collected in Step 1, construct a state matrix D based on SBM.

[0008] Step 3: Obtain the parameter states of the DC power generation power of the photovoltaic panel, the total irradiance on the inclined surface of the photovoltaic panel, the ambient temperature, and the ambient relative humidity when the traveling-wave electric curtain actually operates, and predict the parameter states during normal operation under the current working conditions based on the actual operation parameter states and the state matrix D in Step 2.

[0009] Step 4: According to the actual operating parameter status of the traveling wave electric curtain obtained and the predicted parameter status during normal operation, determine the abnormality factor vectors of the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity respectively. Judge the equipment operating status according to the abnormality factor vectors, and adaptively control the traveling wave electric curtain dust removal equipment according to the equipment operating status.

[0010] Further, the state matrix D constructed in step 2 is to select j groups of typical data from the parameter set X:

[0011]

[0012] Among them, a column vector represents 4 key parameter values during normal operation, that is, X(1) = [x 11 , x 21 , x 31 , x 41 represents the 4 key parameter values during the first group of normal operation, and x 11 , x 21 , x 31 , x 41 represent the parameter values of the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity of the first group. X(2) = [x 12 , x 22 , x 32 , x 42 represents the 4 key parameter values during the second group of normal operation, and x 12 ~x 42 represent the parameter values of the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity of the second group. X(j) = [x 1j , x 2j , x 3j , x 4j represents the 4 key parameter values during the second group of normal operation, and x 1j ~x 4j represent the parameter values of the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity of the jth group.

[0013] Further, the prediction of the parameter status during normal operation under the current working condition in step 3 is specifically as follows:

[0014] Compare the similarity between the actual operating parameter status and the state matrix D to obtain the similarity vector A:

[0015]

[0016] Obtain the weight vector p from the similarity vector:

[0017] p0 = G -1 A

[0018]

[0019]

[0020] Among them, x normal represents the parameter state during normal operation under the predicted current working condition, and x now = [x now,1 x now,2 x now, 3x now,4 T represents the 4-parameter state vector during the current actual operation, and x now,1 x now,2 x now,3 x now,4 respectively represent the DC power generation of the photovoltaic panel, the total irradiance on the inclined surface of the photovoltaic panel, the ambient temperature, and the ambient relative humidity; i represents the number of key parameters, with a value range of 1 - 4, p0 represents the correlation between the currently judged key parameter and the parameter matrix under the normal operation state, and p 0,i represents the 1 - 4 key parameters in the weight vector p0, and p 0,1 refers to the parameter correlation of the DC power generation of the photovoltaic panel, and p 0,2 refers to the parameter correlation of the total irradiance on the inclined surface of the photovoltaic panel, and p 0,3 refers to the parameter correlation of the ambient temperature, and p 0,4 refers to the parameter correlation of the ambient relative humidity.

[0021] Furthermore, the abnormality factor vector in step 4 is specifically:

[0022]

[0023] d = [d1 d2 d3 d4]

[0024] Among them, i takes 1, 2, 3, 4, x now,i represents the parameter state vector during the current actual operation, and x normal,i represents the parameter state during normal operation under the predicted current working condition. The "abnormality" corresponds to the degree of dust accumulation and is the normal state closest to x now obtained by interpolation from the ideal state included in D. The closer x normal is to x now , the more similar x now is to D, and the more normal the equipment operation state is at this time. The greater the deviation between the two, the more abnormal the equipment operation state is at this time.

[0025] ​Further, in step 4, the dust removal device is adaptively controlled according to the device operation status, specifically as follows:

[0026] S5.1: Set the abnormality threshold d0; the target excitation voltage is u;

[0027] S5.2: Calculate the average relative error a of the abnormality degree by comparing with d0:

[0028]

[0029] S5.3: Take 500V as the reference voltage and adjust the target excitation voltage u using the relative error a:

[0030] When the relative error a is greater than 0,

[0031] u(t) = 500|cos(1 + a)| - 200e 0.5t

[0032] When the relative error a is less than 0,

[0033]

[0034] where t is the time for adjusting the voltage;

[0035] S5.4: When the thresholds of the device abnormality degrees d1, d2, d3, and d4 are all less than d0, the dust removal ends.

[0036] Further, the traveling wave electric curtain dust removal device is installed above the solar panel, and it is excited by a three-phase sinusoidal AC power supply. The three-phase sinusoidal alternating current is applied to the transparent electrode, and the transparent electrode is etched on the substrate. An insulating layer is laid above the transparent electrode; the three-phase sinusoidal AC power supply is powered by the solar panel, converted into three-phase alternating current by an inverter, and after voltage regulation control of the three-phase alternating current output by the inverter, it is converted into an AC power supply with adjustable voltage.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] 1. The designed electric curtain excitation power supply comes from the solar panel, realizing on-site power utilization, reducing the amount of cable used, and reducing the energy loss caused by power conversion;

[0039] 2. Single-phase alternating current generates a standing wave electric field. Particles only move back and forth under the action of the standing wave electric field and will only leave the surface of the electric curtain when parameters such as voltage and frequency are coupled to certain conditions. However, the parameter design and control lack theoretical guidance and have great blindness. This design uses three-phase alternating current. When the three-phase AC excitation is turned on, a traveling wave is generated between the electrodes. The charged particles on the traveling wave electric curtain will move in a directional manner along or against the propagation direction of the traveling wave electric field under the action of the traveling wave electric field and are more easily removed;

[0040] 3. The SBM algorithm used in this design is a non-parametric empirical modeling technology. It estimates the state that the equipment should present in its current normal operation based on the state of the equipment in its historical normal operation. It is a weak supervision method that does not require historical parameters to have dust accumulation labels. It dynamically associates the dust accumulation degree with the power generation efficiency, indirectly infers the dust accumulation trend through multivariate efficiency analysis, obtains the equipment abnormality factor, and adjusts the excitation voltage through error analysis to reduce working energy consumption. In addition, it can adapt to occasions with different environmental humidity and different dust removal requirements, and improve the application value of electric curtain dust removal. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the electric curtain structure of the present invention;

[0042] Figure 2 It is the flow chart of SBM algorithm;

[0043] Figure 3 is a control flow chart of the present invention;

[0044] Figure 4 The figure shows a comparison of two voltage outputs during dust removal in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0046] The present invention provides a traveling wave electric curtain adaptive control method based on SBM dust prediction, wherein a traveling wave electric curtain dust removal device is installed above a solar cell panel, wherein the traveling wave electric curtain is a traveling wave electric curtain excited by a three-phase sinusoidal alternating current power supply, wherein the three-phase sinusoidal alternating current power is applied to a transparent electrode so that sunlight can be transmitted to the surface of the solar cell panel, wherein the transparent electrode is etched on a substrate, and an insulating layer is laid above the transparent electrode.

[0047] The three-phase sinusoidal AC excitation power is provided by the solar panel and converted into three-phase AC power by the inverter. The three-phase AC power output by the inverter is converted into an AC power supply with adjustable voltage after voltage regulation control.

[0048] The adaptive control steps of the present invention are:

[0049] 1. Prediction of dust accumulation based on SBM:

[0050] S1. State matrix construction:

[0051] 1) Based on the environmental conditions and performance requirements, four key parameters are selected: the DC power generation of the photovoltaic panels, the total irradiance of the photovoltaic panel slope, the ambient temperature and the ambient relative humidity.

[0052] 2) Collect the above four key parameter values when the device was operating normally (without dust accumulation) in the past to form the standard parameter set X.

[0053] 3) Select j groups of typical data from the parameter set X to construct the state matrix D:

[0054]

[0055] Among them, a column vector represents the four key parameter values during a group of normal operations, that is, X(1) = [x 11 , x 21 , x 31 , x 41 represents the four key parameter values during the first group of normal operations, and x 11 , x 21 , x 31 , x 41 represent the first group of photovoltaic panel DC power generation, total irradiance on the inclined surface of the photovoltaic panel, ambient temperature, and ambient relative humidity parameter values. X(2) = [x 12 , x 22 , x 32 , x 42 represents the four key parameter values during the second group of normal operations, and x 12 ~ x 42 represent the second group of photovoltaic panel DC power generation, total irradiance on the inclined surface of the photovoltaic panel, ambient temperature, and ambient relative humidity parameter values. X(j) = [x 1j , x 2j , x 3j , x 4j represents the four key parameter values during the jth group of normal operations, and x 1j ~ x 4j represent the jth group of photovoltaic panel DC power generation, total irradiance on the inclined surface of the photovoltaic panel, ambient temperature, and ambient relative humidity parameter values.

[0056] S2. Obtain the actual operating parameter status:

[0057] x now = [x now,1 x now,2 x now,3 x now,4 T

[0058] S3. Predict the parameter status x normal .

[0059] Compare the similarity between the actual operating parameter status and the state matrix D to obtain the similarity vector A: ​

[0060]

[0061] Obtain the weight vector p from the similarity vector:

[0062] p0 = G -1 A

[0063]

[0064] where x normal represents the parameter state during normal operation under the predicted current working condition, and x now = [x now,1 x now,2 x now, 3x now,4 T represents the 4-parameter state vector during the current actual operation. x now,1 x now,2 x now,3 x now,4 respectively represent the DC power generation of the photovoltaic panel, the total irradiance on the inclined surface of the photovoltaic panel, the ambient temperature, and the ambient relative humidity; i represents the number of key parameters, with a value range of 1 - 4, p0 represents the correlation between the currently judged key parameter and the parameter matrix under the normal operation state, and p 0,i represents the 1 - 4 key parameters in the weight vector p0, p 0,1 refers to the parameter correlation of the DC power generation of the photovoltaic panel, p 0,2 refers to the parameter correlation of the total irradiance on the inclined surface of the photovoltaic panel, p 0,3 refers to the parameter correlation of the ambient temperature, p 0,4 refers to the parameter correlation of the ambient relative humidity.

[0065] S4. Obtain the device abnormality factor vector:

[0066]

[0067] d = [d1 d2 d3 d4]

[0068] The "abnormality" corresponds to the dust accumulation degree of the component, and it is the normal state closest to x now obtained by interpolation from the ideal state included in D. The closer x normal is to x now , the more similar x now is to D, and the more normal the device operation state is at this time. The greater the deviation between the two, the more abnormal the device operation state is at this time. Since the selected parameters include various environmental parameters that affect the power generation, if equipment depreciation is not considered, the main cause of the deviation is the dust accumulation degree.

[0069] II. Voltage regulation control method:​

[0070] S1. Set the abnormality threshold d0.

[0071] S2. The target excitation voltage is u.

[0072] By comparing with d0, calculate the average relative error a of the abnormality degree:

[0073]

[0074] Take 500V as the reference voltage and adjust the target voltage u using the relative error a:

[0075] When the relative error a is greater than 0,

[0076] u(t) = 500|cos(1 + a)| - 200e 0.5t

[0077] When the relative error a is less than 0,

[0078]

[0079] where t is the time for adjusting the voltage.

[0080] S3. When the abnormality degrees d1, d2, d3, d4 thresholds of the device are all less than d0, the dust removal ends.

[0081] The present invention uses the SBM model to monitor the dust accumulation degree of the photovoltaic panel, and based on this, designs a more energy-saving and efficient adaptive control method for the dust removal voltage of the photovoltaic panel.

[0082] The voltage regulation effect is as follows:

[0083] In the experiment, the distance between the electrode plates of the made electric curtain is selected to be 0.5mm, and several standing wave electric curtains with the same structural parameters and in a clean state are used as the test plates to be dust-removed after being naturally contaminated for one week. The energized voltage during the dust removal of the photovoltaic panel is detected respectively under the two modes of the given voltage and the adjustable voltage (the present invention), and the output values of the voltage within a cleaning cycle can be obtained as follows Figure 4 as shown.

[0084] By comparing the voltage output curves, it can be obtained that when the SBM dust accumulation monitoring model is not adopted, in order to achieve the dust removal effect, the photovoltaic panel needs to continuously connect a 500V voltage to meet the maximum dust removal power. When the SBM dust accumulation prediction model is adopted, the voltage connected to the photovoltaic panel can detect the dust accumulation amount at any time, perform efficient automatic dust removal in a short time, and will automatically cut off the power when the dust removal threshold is reached, saving energy consumption and achieving the effect of energy conservation and emission reduction.

Claims

1. A method for adaptive control of a traveling wave electric curtain based on SBM dust prediction, characterized in that It includes the following steps: Step 1: Collect the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity during the normal operation of the traveling-wave electric curtain dust removal equipment in history, and form a standard parameter set X. The normal operation of the equipment refers to the operation process of the traveling-wave electric curtain without dust accumulation; Step 2: Based on the standard parameter set X collected in Step 1, construct a state matrix D based on SBM; Step 3: Obtain the parameter states of the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity during the actual operation of the traveling-wave electric curtain, and predict the parameter states during normal operation under the current working conditions based on the actual operation parameter states and the state matrix D in Step 2; Step 4: According to the obtained actual operation parameter states of the traveling-wave electric curtain and the predicted parameter states during normal operation, respectively determine the abnormality factor of the DC power generation of the photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity. Judge the equipment operation state according to the abnormality factor, and adaptively control the traveling-wave electric curtain dust removal equipment according to the equipment operation state.

2. The adaptive control method for traveling wave electric curtain based on SBM dust prediction according to claim 1, wherein In Step 2, when constructing the state matrix D, j sets of typical data are selected from the parameter set X: Among them, a column vector represents a set of 4 key parameter values during normal operation, that is, X(1) = [x 11 , x 21 , x 31 , x 41 represents the 4 key parameter values of the first set during normal operation, and x 11 , x 21 , x 31 , x 41 represent the parameter values of the DC power generation of the first set of photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity. X(2) = [x 12 , x 22 , x 32 , x 42 represents the 4 key parameter values of the second set during normal operation, and x 12 ~x 42 represent the parameter values of the DC power generation of the second set of photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity. X(j) = [x 1j , x 2j , x 3j , x 4j represents the 4 key parameter values of the j-th set during normal operation, and x 1j ~x 4j represent the parameter values of the DC power generation of the j-th set of photovoltaic panels, the total irradiance on the inclined surface of the photovoltaic panels, the ambient temperature, and the ambient relative humidity.

3. The adaptive control method of the traveling wave electric curtain based on SBM dust prediction according to claim 1, characterized in that In Step 3, the specific method for predicting the parameter states during normal operation under the current working conditions is as follows: Compare the similarity between the actual operation parameter states and the state matrix D to obtain a similarity vector A: Obtain a weight vector p from the similarity vector: p0 = G -1 A where x normal represents the parameter state during normal operation under the predicted current working condition x now = [x now,1 x now,2 x now,3 x now,4 T represents the state vector of 4 parameters during the current actual operation. x now,1 x now, 2x now,3 x now,4 respectively represent the DC power generation of the photovoltaic panel, the total irradiance on the inclined surface of the photovoltaic panel, the ambient temperature, and the ambient relative humidity; i represents the number of key parameters, with a value range of 1 - 4, p0 represents the correlation between the currently judged key parameter and the parameter matrix under normal operating conditions, p 0,i represents 1 - 4 key parameters in the weight vector p0, p 0,1 refers to the parameter correlation of the DC power generation of the photovoltaic panel, p 0,2 refers to the parameter correlation of the total irradiance on the inclined surface of the photovoltaic panel, p 0,3 refers to the parameter correlation of the ambient temperature, p 0,4 refers to the parameter correlation of the ambient relative humidity.​ 4. The adaptive control method of the traveling wave electric curtain based on SBM dust prediction according to claim 1, characterized in that, The abnormality factor vector in Step 4 is specifically: d = [d1 d2 d3 d4] Among them, i takes values of 1, 2, 3, and 4, and x now,i represents the parameter state vector during the current actual operation, and x normal,i represents the parameter state during the predicted normal operation under the current working condition. The "abnormality degree" corresponds to the ash fouling degree and is the normal state closest to the ideal state contained in D obtained by interpolation with x now . The closer x normal is to x now , the more similar x now is to D, and the more normal the operating state of the equipment is at this time. The greater the deviation between the two, the more abnormal the operating state of the equipment is at this time.

5. The adaptive control method of the traveling wave electric curtain based on SBM dust prediction according to claim 4, wherein In Step 4, adaptively controlling the dust removal equipment according to the equipment operation state is specifically: S5.1: Set an abnormality threshold d0; the target excitation voltage is u; S5.2: Calculate the average relative error a of the abnormality by comparing with d0: S5.3: Take 500V as the reference voltage and adjust the target excitation voltage u using the relative error a: When the relative error a is greater than 0, u(t) = 500|cos(1 + a)| - 200e 0.5t When the relative error a is less than 0, where t is the time for adjusting the voltage; S5.4: When the abnormality thresholds d1, d2, d3, and d4 of the equipment are all less than d0, the dust removal ends.

6. The adaptive control method for traveling wave electric curtain based on SBM dust prediction according to any one of claims 1 to 5, characterized in that, The traveling-wave electric curtain dust removal equipment is installed above the solar panels, and it is excited by a three-phase sinusoidal AC power supply. The three-phase sinusoidal alternating current is applied to the transparent electrode, and the transparent electrode is etched on the substrate. An insulating layer is laid above the transparent electrode; the three-phase sinusoidal AC power supply is powered by the solar panels, and after being converted by an inverter, it becomes three-phase alternating current. After voltage regulation control of the three-phase alternating current output by the inverter, it is converted into an AC power supply with adjustable voltage.

Images