Variable frequency control method and system of heating, ventilation and air conditioning coupled with industrial waste heat

Abstract

The application provides a variable frequency control method and system of a heating, ventilation and air conditioning coupled with industrial waste heat. The trajectory extension coefficient is determined by collecting indoor temperature, suction temperature, exhaust temperature, suction pressure, exhaust pressure, industrial waste heat source temperature and medium flow, combining with the frequency adjustment amount of the last control cycle to construct a target compression process trajectory line spliced by Bezier curves. The waste heat potential energy factor is calculated according to the heat source temperature, temperature change rate and medium flow, and the suction working condition, waste heat potential energy factor and operating frequency are input into the adjustable first-order multivariable grey model to predict the exhaust temperature of the next cycle. The first frequency adjustment amount is generated according to the indoor temperature difference, the second frequency adjustment amount is generated according to the exhaust temperature trajectory deviation, and the variable frequency command is fused and output according to the waste heat potential energy factor. When the waste heat potential energy factor is not less than the weight fusion switching threshold, the exhaust side trajectory adjustment is emphasized, and when the threshold is lower, the indoor temperature control is emphasized, and the waste heat utilization and indoor constant temperature are considered.

Description

Technical Field

[0001] This application belongs to the field of control, and in particular relates to a variable frequency control method and system for HVAC that couples industrial waste heat. Background Technology

[0002] Heating, ventilation, and air conditioning (HVAC) systems are major energy-consuming equipment in buildings and industrial settings, and improving their operational efficiency is crucial for reducing energy consumption. Variable frequency drive (VFD) technology, which dynamically adjusts equipment operation based on actual load demand, has been widely applied in HVAC systems. To further reduce overall system energy consumption and improve energy cascade utilization, waste heat generated during industrial production can be introduced into the HVAC system as an auxiliary heat source. By recovering and utilizing industrial waste heat, not only can traditional energy consumption be reduced, but the environmental heat load caused by waste heat emissions can also be lowered. Therefore, HVAC systems integrating VFD control and waste heat recovery have significant economic and ecological benefits in industrial parks, large commercial complexes, and regional centralized energy supply projects.

[0003] However, industrial waste heat is affected by production processes, equipment loads, and operating cycles, resulting in significant time-varying and uncertainties in heat source temperature and medium flow rate. Traditional variable frequency control strategies often rely primarily on indoor temperature deviations or only provide delayed feedback on heat source fluctuations, making it difficult to establish a dynamic coupling relationship between the compressor's thermodynamic process and waste heat potential energy changes. When waste heat parameters change abruptly, existing control methods struggle to predict and adjust the compressor's operating state in a timely manner, easily leading to exhaust temperature fluctuations and deviations from the target operating trajectory during compression, thus affecting unit lifespan and operational reliability. Furthermore, existing mechanisms cannot flexibly allocate control weights between indoor temperature control and waste heat utilization based on the strength of waste heat potential energy, making it difficult to achieve comprehensive optimization of system energy efficiency while ensuring stable indoor temperature. Summary of the Invention

[0004] In order to simultaneously improve the reliability of unit operation and the overall energy efficiency of the system, this invention provides a variable frequency control method and system for HVAC that couples industrial waste heat.

[0005] The first aspect of this invention provides a variable frequency control method for HVAC systems coupled with industrial waste heat, comprising the following steps:

[0006] The system acquires indoor temperature, suction pressure, discharge pressure, suction temperature, discharge temperature, industrial waste heat source temperature, and medium flow rate. Combined with the trajectory extension coefficient determined by the second frequency adjustment of the previous control cycle, a target compression process trajectory line spliced ​​with Bézier curves is constructed based on the suction pressure and the industrial waste heat source temperature. The control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature.

[0007] The waste heat potential energy factor is calculated based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the medium flow rate. The exhaust temperature of the next cycle is predicted by a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor, with the intake state, the waste heat potential energy factor, and the current operating frequency as inputs.

[0008] The control direction of the indoor temperature difference is determined according to the current operating mode, and the first frequency adjustment amount is calculated according to the patterned temperature difference between the target indoor temperature and the indoor temperature. The second frequency adjustment amount is calculated according to the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line. The first frequency adjustment amount and the second frequency adjustment amount are weighted and fused based on the waste heat potential energy factor to obtain the frequency command. When the waste heat potential energy factor is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is increased. When the waste heat potential energy factor is lower than the weight fusion switching threshold, the weight of the first frequency adjustment amount is increased.

[0009] Optionally, the step of constructing a target compression process trajectory line by splicing Bézier curves based on the suction pressure and the industrial waste heat source temperature, wherein the control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature, includes:

[0010] Obtain the preset maximum upper limit exhaust pressure, and subtract the current intake pressure to get the pressure range difference;

[0011] The pressure interval difference is divided into three parts, and the pressure coordinates of the starting endpoint, two segment splicing nodes and the ending endpoint are determined. The pressure coordinates of two internal control points are generated between each adjacent endpoint according to a preset ratio, forming the segment endpoint and internal control point coordinate reference required for three third-order Bézier curves.

[0012] The refrigerant suction state entropy value is determined based on the current suction pressure and current suction temperature, and the corresponding isentropic compression theoretical temperature is calculated based on the suction state entropy value and each pressure coordinate reference, which serves as the base temperature for the corresponding control point.

[0013] Obtain the refrigerant saturation temperature corresponding to the current suction pressure, and calculate the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source.

[0014] The temperature coordinate offset of the control point is obtained by multiplying the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source by a proportional constant and combining it with the trajectory extension coefficient.

[0015] The temperature coordinates are obtained by adding the base temperature corresponding to each pressure coordinate reference with the temperature coordinate offset. The pressure coordinates of the segment endpoints, splicing nodes and internal control points are combined with the corresponding temperature coordinates to generate the control point coordinates of three third-order Bézier curves. The three Bézier curves are generated by the control point coordinates and smoothly spliced ​​to output the target compression process trajectory line.

[0016] Optionally, the step of calculating the waste heat potential energy factor based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the medium flow rate includes:

[0017] Calculate the temperature deviation obtained by subtracting the preset reference temperature from the current temperature of the industrial waste heat source, and perform non-negative truncation processing on the temperature deviation.

[0018] The absolute value of the time-varying rate of temperature change of the industrial waste heat source within the current control cycle is calculated as the temperature fluctuation rate.

[0019] The dimensionless flow coefficient is obtained by obtaining the flow rate of the medium and dividing it by the rated flow rate reference.

[0020] The temperature deviation and the temperature fluctuation rate are respectively divided by the corresponding preset reference values ​​to perform dimensionless processing;

[0021] The product of the dimensionless flow coefficient and the dimensionless temperature deviation is used as the numerator, and the sum of the dimensionless temperature fluctuation rate and the preset constant is used as the denominator to calculate the ratio. The ratio is multiplied by the potential energy constant to obtain the original value of waste heat potential energy. The original value of waste heat potential energy is normalized according to the minimum and maximum original values ​​in the historical calibration samples, and the normalization result is limited to the range of 0 to 1 to obtain the waste heat potential energy factor.

[0022] Optionally, the step of predicting the exhaust temperature of the next cycle using a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor includes:

[0023] Determine whether the value of the waste heat potential energy factor is not less than the model parameter switching threshold;

[0024] If the value is not less than the model parameter switching threshold, then the background value is constructed using an integral formula, and the development coefficient is set to the first preset development coefficient.

[0025] If the value is less than the model parameter switching threshold, the background value is constructed using a weighted average formula based on the three most recent data, and the development coefficient is set to the second preset development coefficient.

[0026] Wherein, the first preset development coefficient is less than the second preset development coefficient;

[0027] Using the first-order multivariate grey prediction model after adjusting the background value and the development coefficient, while keeping the development coefficient unchanged, only the grey action and the driving coefficients corresponding to each input variable are solved, and the exhaust temperature of the next cycle is output based on the solution results.

[0028] Optionally, the step of weighting and fusing the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor to obtain the frequency command includes:

[0029] Determine whether the waste heat potential energy factor is not less than the weight fusion switching threshold;

[0030] When the weight is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is set to 0.8, and the weight of the first frequency adjustment amount is set to 0.2;

[0031] When the weight is less than the weighted fusion switching threshold, the weight of the first frequency adjustment is set to 0.8, and the weight of the second frequency adjustment is set to 0.2.

[0032] The first frequency adjustment amount and the second frequency adjustment amount are multiplied by their respective weights and then added together to obtain the comprehensive frequency increment. The frequency command is then generated based on the comprehensive frequency increment and the current operating frequency.

[0033] Optionally, calculating the second frequency adjustment amount based on the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line includes:

[0034] Find a target trajectory point on the target compression process trajectory line that corresponds to the current exhaust pressure with pressure coordinates, the target trajectory point including pressure coordinates and temperature coordinates;

[0035] Extract the temperature coordinates of the target trajectory points as the target exhaust temperature;

[0036] The exhaust temperature deviation is obtained by subtracting the next cycle exhaust temperature from the target exhaust temperature.

[0037] The basic adjustment amount is obtained by multiplying the exhaust temperature deviation by a preset proportional constant of 0.6.

[0038] The basic adjustment amount and the preset feedforward compensation amount are superimposed with a limited amplitude, and the direction of the superimposed second frequency adjustment amount is constrained to be consistent with the adjustment direction corresponding to the exhaust temperature deviation, so as to obtain the second frequency adjustment amount.

[0039] A second aspect of the present invention provides a variable frequency control system for HVAC coupled with industrial waste heat, comprising the following modules:

[0040] The module is used to acquire indoor temperature, suction pressure, discharge pressure, suction temperature, discharge temperature, industrial waste heat source temperature and medium flow rate. Combined with the trajectory extension coefficient determined by the second frequency adjustment of the previous control cycle, the target compression process trajectory line spliced ​​by Bézier curves is constructed based on the suction pressure and the industrial waste heat source temperature. The control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature.

[0041] The adjustment module is used to calculate the waste heat potential energy factor based on the industrial waste heat source temperature, the heat source temperature change rate and the medium flow rate. Taking the intake state, the waste heat potential energy factor and the current operating frequency as input, the exhaust temperature of the next cycle is predicted by a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor.

[0042] The calculation module is used to determine the control direction of the indoor temperature difference according to the current operating mode, calculate the first frequency adjustment amount according to the patterned temperature difference between the target indoor temperature and the indoor temperature, calculate the second frequency adjustment amount according to the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line, and obtain a frequency command by weighted fusion of the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor. When the waste heat potential energy factor is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is increased; when the waste heat potential energy factor is lower than the weight fusion switching threshold, the weight of the first frequency adjustment amount is increased.

[0043] Preferably, the step of constructing a target compression process trajectory line by splicing Bézier curves based on the suction pressure and the industrial waste heat source temperature, wherein the control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature, includes:

[0044] Obtain the preset maximum upper limit exhaust pressure, and subtract the current intake pressure to get the pressure range difference;

[0045] The pressure interval difference is divided into three parts, and the pressure coordinates of the starting endpoint, two segment splicing nodes and the ending endpoint are determined. The pressure coordinates of two internal control points are generated between each adjacent endpoint according to a preset ratio, forming the segment endpoint and internal control point coordinate reference required for three third-order Bézier curves.

[0046] The refrigerant suction state entropy value is determined based on the current suction pressure and current suction temperature, and the corresponding isentropic compression theoretical temperature is calculated based on the suction state entropy value and each pressure coordinate reference, which serves as the base temperature for the corresponding control point.

[0047] Obtain the refrigerant saturation temperature corresponding to the current suction pressure, and calculate the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source.

[0048] The temperature coordinate offset of the control point is obtained by multiplying the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source by a proportional constant and combining it with the trajectory extension coefficient.

[0049] The temperature coordinates are obtained by adding the base temperature corresponding to each pressure coordinate reference with the temperature coordinate offset. The pressure coordinates of the segment endpoints, splicing nodes and internal control points are combined with the corresponding temperature coordinates to generate the control point coordinates of three third-order Bézier curves. The three Bézier curves are generated by the control point coordinates and smoothly spliced ​​to output the target compression process trajectory line.

[0050] Preferably, the step of calculating the waste heat potential energy factor based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the medium flow rate includes:

[0051] Calculate the temperature deviation obtained by subtracting the preset reference temperature from the current temperature of the industrial waste heat source, and perform non-negative truncation processing on the temperature deviation.

[0052] The absolute value of the time-varying rate of temperature change of the industrial waste heat source within the current control cycle is calculated as the temperature fluctuation rate.

[0053] The dimensionless flow coefficient is obtained by obtaining the flow rate of the medium and dividing it by the rated flow rate reference.

[0054] The temperature deviation and the temperature fluctuation rate are respectively divided by the corresponding preset reference values ​​to perform dimensionless processing;

[0055] The product of the dimensionless flow coefficient and the dimensionless temperature deviation is used as the numerator, and the sum of the dimensionless temperature fluctuation rate and the preset constant is used as the denominator to calculate the ratio. The ratio is multiplied by the potential energy constant to obtain the original value of waste heat potential energy. The original value of waste heat potential energy is normalized according to the minimum and maximum original values ​​in the historical calibration samples, and the normalization result is limited to the range of 0 to 1 to obtain the waste heat potential energy factor.

[0056] Preferably, the step of predicting the exhaust temperature of the next cycle using a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor includes:

[0057] Determine whether the value of the waste heat potential energy factor is not less than the model parameter switching threshold;

[0058] If the value is not less than the model parameter switching threshold, then the background value is constructed using an integral formula, and the development coefficient is set to the first preset development coefficient.

[0059] If the value is less than the model parameter switching threshold, the background value is constructed using a weighted average formula based on the three most recent data, and the development coefficient is set to the second preset development coefficient.

[0060] Wherein, the first preset development coefficient is less than the second preset development coefficient;

[0061] Using the first-order multivariate grey prediction model after adjusting the background value and the development coefficient, while keeping the development coefficient unchanged, only the grey action and the driving coefficients corresponding to each input variable are solved, and the exhaust temperature of the next cycle is output based on the solution results.

[0062] Preferably, the step of weighting and fusing the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor to obtain the frequency command includes:

[0063] Determine whether the waste heat potential energy factor is not less than the weight fusion switching threshold;

[0064] When the weight is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is set to 0.8, and the weight of the first frequency adjustment amount is set to 0.2;

[0065] When the weight is less than the weighted fusion switching threshold, the weight of the first frequency adjustment is set to 0.8, and the weight of the second frequency adjustment is set to 0.2.

[0066] The first frequency adjustment amount and the second frequency adjustment amount are multiplied by their respective weights and then added together to obtain the comprehensive frequency increment. The frequency command is then generated based on the comprehensive frequency increment and the current operating frequency.

[0067] Preferably, the step of calculating the second frequency adjustment amount based on the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line includes:

[0068] Find a target trajectory point on the target compression process trajectory line that corresponds to the current exhaust pressure with pressure coordinates, the target trajectory point including pressure coordinates and temperature coordinates;

[0069] Extract the temperature coordinates of the target trajectory points as the target exhaust temperature;

[0070] The exhaust temperature deviation is obtained by subtracting the next cycle exhaust temperature from the target exhaust temperature.

[0071] The basic adjustment amount is obtained by multiplying the exhaust temperature deviation by a preset proportional constant of 0.6.

[0072] The basic adjustment amount and the preset feedforward compensation amount are superimposed with a limited amplitude, and the direction of the superimposed second frequency adjustment amount is constrained to be consistent with the adjustment direction corresponding to the exhaust temperature deviation, so as to obtain the second frequency adjustment amount.

[0073] This invention collects multi-dimensional operating parameters and industrial waste heat parameters to construct a target compression process trajectory line formed by smoothly splicing Bezier curves. This makes the compressor's operating path planning more continuous and stable, thereby improving the reliability of the waste heat coupling utilization process. The waste heat potential energy factor is calculated based on the waste heat source temperature, the rate of change of the heat source temperature, and the medium flow rate. A first-order multivariate grey prediction model is used to predict the exhaust temperature of the next cycle in advance, improving the system's ability to predict and adjust for waste heat fluctuations and external heat source disturbances. Simultaneously, a weighted fusion of the first frequency adjustment quantity based on the indoor temperature difference and the second frequency adjustment quantity based on the compression trajectory deviation is performed, and the weights of the two types of adjustment quantities are dynamically allocated according to the waste heat potential energy factor. When the waste heat potential energy is sufficient, the focus is on adjusting the compression trajectory and exhaust side state to promote deep recovery of industrial waste heat and reduce energy consumption; when the waste heat potential energy is insufficient, the focus is on indoor temperature regulation to ensure indoor environmental stability. This method can balance indoor comfort, waste heat utilization efficiency, and unit operating reliability, improving the overall energy efficiency of the HVAC system. Attached Figure Description

[0074] Figure 1 A flowchart of a variable frequency control method for HVAC systems coupled with industrial waste heat;

[0075] Figure 2 This is a schematic diagram of the dynamic change curve of the waste heat potential energy factor.

[0076] Figure 3 This is a schematic diagram of the variable frequency control weight switching characteristics. Detailed Implementation

[0077] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this specification as detailed in the appended claims.

[0078] It should be understood that the terms “comprising” and “having”, and any variations thereof, in the embodiments of this specification are intended to cover but not exclude inclusion. For example, a product or device that includes a series of components is not necessarily limited to those components that are explicitly listed, but may include other components that are not explicitly listed or that are inherent to such product or device.

[0079] Example 1

[0080] In Embodiment 1 of the present invention, as Figure 1 As shown, a variable frequency control method for HVAC systems coupled with industrial waste heat includes the following steps:

[0081] S1: Obtain indoor temperature, suction pressure, discharge pressure, suction temperature, discharge temperature, industrial waste heat source temperature, and medium flow rate. Combined with the trajectory extension coefficient determined by the second frequency adjustment amount of the previous control cycle, construct a target compression process trajectory line spliced ​​with Bézier curves based on the suction pressure and the industrial waste heat source temperature. The control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature.

[0082] A platinum resistance temperature sensor is used to collect indoor temperature, intake temperature, exhaust temperature, and industrial waste heat source temperature. A piezoresistive pressure transmitter is used to collect intake pressure and exhaust pressure. An electromagnetic flowmeter is used to collect medium flow rate. These sensors convert analog signals into digital signals and transmit them to a microcontroller via Modbus communication protocol based on an RS485 hardware interface. The second frequency adjustment value of the previous control cycle is extracted from the microcontroller's memory and input into a preset linear mapping function to calculate the trajectory extension coefficient used to stretch or scale the curve geometry. The trajectory extension coefficient is obtained through amplitude-limited linear mapping, preferably... ,in This represents the second frequency adjustment value from the previous control cycle, where k is a preset mapping coefficient. and These are the preset lower and upper limits, respectively. Greater than 0. Using the REFPROP property calculation core algorithm library provided by the National Institute of Standards and Technology (NIST), the refrigerant saturation temperature is obtained by querying the current suction pressure. The algebraic difference between the refrigerant saturation temperature and the industrial waste heat source temperature is calculated as the temperature difference reference quantity. This algebraic difference is calculated in the direction of "refrigerant saturation temperature minus industrial waste heat source temperature" and multiplied by a preset negative proportionality constant. Therefore, when the industrial waste heat source temperature increases, the absolute value of the temperature difference reference quantity increases, and the temperature offset increases along a preset upward direction, avoiding inconsistencies in the direction of the temperature offset.

[0083] The coordinate point formed by the intake pressure and intake temperature is used as the starting endpoint of the target compression process trajectory line, and the coordinate point formed by the preset maximum upper limit exhaust pressure and the empirically highest exhaust temperature at that pressure is used as the ending endpoint of the target compression process trajectory line. The current exhaust pressure is not used as the ending endpoint, but is used in S3 to find the target temperature along the constructed trajectory line. The temperature offset is obtained by multiplying the temperature difference reference amount by a preset negative proportionality constant and the trajectory extension coefficient. Segment nodes are determined by dividing the pressure interval into three equal parts between the starting endpoint and the ending endpoint. For three segments of third-order Bézier curves, the first segment consists of the starting endpoint, two internal control points, and the first splicing node; the second segment consists of the first splicing node, two internal control points, and the second splicing node; the third segment consists of the second splicing node, two internal control points, and the ending endpoint. Thus, the three segments of the curve include a total of four endpoints or splicing nodes and six internal control points, which meets the number of control points required for splicing three segments of third-order Bézier curves. A preset Bézier curve calculation program is called in the microcontroller or the host control program, or the third-order Bézier curve is generated using the de Castelli algorithm. The de Castellio algorithm is a method for calculating the stability of arbitrary parameter positions on a Bézier curve by recursively performing layer-by-layer linear interpolation on the control points of the curve. For cases with large operating conditions, multiple third-order Bézier curve segments are joined end-to-end. At the joining nodes, the line connecting the last two control points of the preceding curve segment is made collinear with and in the same direction as the line connecting the first two control points of the following curve segment, achieving first-order geometric continuity and constructing a complete trajectory line for the target compression process.

[0084] In one embodiment, the step of constructing a target compression process trajectory line by splicing Bezier curves based on the suction pressure and the industrial waste heat source temperature, wherein the control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature, includes:

[0085] Obtain the preset maximum upper limit exhaust pressure, and subtract the current intake pressure to get the pressure range difference;

[0086] The pressure interval difference is divided into three parts, and the pressure coordinates of the starting endpoint, two segment splicing nodes and the ending endpoint are determined. The pressure coordinates of two internal control points are generated between each adjacent endpoint according to a preset ratio, forming the segment endpoint and internal control point coordinate reference required for three third-order Bézier curves.

[0087] The refrigerant suction state entropy value is determined based on the current suction pressure and current suction temperature, and the corresponding isentropic compression theoretical temperature is calculated based on the suction state entropy value and each pressure coordinate reference, which serves as the base temperature for the corresponding control point.

[0088] Obtain the refrigerant saturation temperature corresponding to the current suction pressure, and calculate the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source.

[0089] The temperature coordinate offset of the control point is obtained by multiplying the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source by a proportional constant and combining it with the trajectory extension coefficient.

[0090] The temperature coordinates are obtained by adding the base temperature corresponding to each pressure coordinate reference with the temperature coordinate offset. The pressure coordinates of the segment endpoints, splicing nodes and internal control points are combined with the corresponding temperature coordinates to generate the control point coordinates of three third-order Bézier curves. The three Bézier curves are generated by the control point coordinates and smoothly spliced ​​to output the target compression process trajectory line.

[0091] Taking R410A refrigerant as an example, the preset maximum upper limit discharge pressure is set to 3.8 MPa. If the current suction pressure collected by the sensor in real time is 0.8 MPa, the difference between the two is 3.0 MPa. This 3.0 MPa is divided into three equal parts, each spanning 1.0 MPa. The pressure coordinates of the two segment splicing nodes and the termination endpoint are determined sequentially, namely 1.8 MPa, 2.8 MPa, and 3.8 MPa. These are combined with the starting endpoint of 0.8 MPa to form the segment endpoint pressure coordinates of the three Bézier curves. Internal control point pressure coordinates are further generated between each adjacent endpoint to complete the construction of the three third-order Bézier curves. The system's refrigerant property parameter database is called. Assuming the current suction temperature is 10℃, the theoretical temperatures corresponding to isentropic compression to the above three pressure coordinate references under this suction state are calculated, yielding 35℃, 55℃, and 75℃ respectively. These temperatures are determined as the base temperatures for each control point. The temperature coordinate offset modulation stage then begins. By consulting the corresponding physical property data table, the refrigerant saturation temperature corresponding to the current suction pressure of 0.8 MPa is found to be approximately 5°C. If the current industrial waste heat source temperature probe feedback value is 45°C, the difference between the two is -40°C. The preset proportional constant is set to -0.1, and the trajectory extension coefficient of the previous control cycle is extracted, such as 1.2. The temperature offset is calculated as 4.8°C through multiplication. This 4.8°C offset is added to the three base temperatures mentioned above to calculate the modulated temperature coordinates as 39.8°C, 59.8°C, and 79.8°C, respectively. These coordinates are then combined with the corresponding pressure coordinates to generate segmented endpoint coordinate sequences (1.8 MPa, 39.8°C), (2.8 MPa, 59.8°C), and (3.8 MPa, 79.8°C). Among them, (1.8MPa, 39.8℃), (2.8MPa, 59.8℃), and (3.8MPa, 79.8℃) are three segment endpoints or splicing nodes, not all control points of the three third-order Bézier curves. The system further generates two internal control points at one-third and two-thirds positions in the three pressure ranges of 0.8MPa to 1.8MPa, 1.8MPa to 2.8MPa, and 2.8MPa to 3.8MPa, respectively. The coordinates of the internal control points are corrected according to the tangential collinearity constraint of adjacent curve segments, thereby generating three smoothly spliced ​​Bézier curves. The temperature coordinates of each internal control point are obtained by superimposing the temperature offset on the isentropic compression theoretical temperature corresponding to its pressure coordinate; or by linear interpolation of the temperature coordinates of adjacent segment endpoints, and then corrected according to the tangential continuity constraint of adjacent curve segments. When correcting the coordinates of the internal control points, the pressure coordinates of each internal control point are constrained to be within the pressure range of the corresponding segment endpoints, and to increase monotonically along the direction from the intake pressure to the maximum upper limit exhaust pressure, so as to ensure that any effective exhaust pressure corresponds to a unique target temperature on the target compression process trajectory line.

[0092] S2, calculate the waste heat potential energy factor based on the industrial waste heat source temperature, the rate of change of heat source temperature and the medium flow rate, and use the intake state, the waste heat potential energy factor and the current operating frequency as inputs to predict the exhaust temperature of the next cycle through a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor.

[0093] The temperature difference of the industrial waste heat source during two consecutive sampling cycles of the microcontroller is calculated and divided by the sampling time interval to obtain the industrial waste heat source temperature change rate. The industrial waste heat source temperature and the medium flow rate together characterize the intensity of waste heat utilization. The absolute value of the industrial waste heat source temperature change rate is used as the suppression factor for waste heat stability, rather than as the positive heat energy compensation factor. Specifically, the static usable heat energy term formed by temperature deviation and medium flow rate is used as the numerator, and the stability suppression term formed by temperature fluctuation rate and a preset constant is used as the denominator to calculate the original ratio of waste heat potential energy. When the industrial waste heat source temperature is lower than the preset reference temperature, the temperature deviation is truncated to zero to indicate that the current heat source does not have an effective waste heat contribution, avoiding negative values ​​for the waste heat potential energy factor. Subsequently, the original ratio is normalized based on the minimum and maximum potential energy values ​​determined by historical calibration samples, and the normalization result is truncated within the range of 0 to 1 to obtain a waste heat potential energy factor with a value between 0 and 1. The intake pressure and intake temperature are combined as a multidimensional feature vector representing the intake state. The waste heat potential energy factor and the current operating frequency are combined as the input feature sequence of the system. The historical exhaust temperature is used as the system behavior sequence to construct a multivariate first-order grey prediction model, namely the GM(1,N) model.

[0094] To adjust the model parameters based on the waste heat potential energy factor, a piecewise adjustment relationship is established between the waste heat potential energy factor, the construction method of the background value of the grey prediction model, and the setting value of the development coefficient. This piecewise adjustment relationship involves constructing the background value using an integral formula and setting the development coefficient to a smaller constant when the waste heat potential energy factor is high, and constructing the background value using a weighted average formula of the three most recent data points and setting the development coefficient to a larger constant when the waste heat potential energy factor is low. When solving the multivariate first-order grey prediction model, an accumulation generation operation is performed on the input feature sequence and the system behavior sequence to obtain an accumulation sequence. A data matrix and a data vector are formed according to the adjusted background value construction method. In this embodiment, the development coefficient has already been set to either a first preset development coefficient or a second preset development coefficient based on the waste heat potential energy factor. Therefore, when calling the lstsq least squares function in the linalg submodule, only the grey action and the remaining parameters corresponding to each input variable are solved, without repeatedly solving for the development coefficient. The obtained parameters are used to construct a discrete-time response function to calculate the accumulated predicted value. A cumulative subtraction and restoration operation is performed on the accumulated predicted value to output the predicted exhaust temperature for the next control cycle.

[0095] In one embodiment, the calculation of the waste heat potential energy factor based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the medium flow rate includes:

[0096] Calculate the temperature deviation obtained by subtracting the preset reference temperature from the current temperature of the industrial waste heat source, and perform non-negative truncation processing on the temperature deviation.

[0097] The absolute value of the time-varying rate of temperature change of the industrial waste heat source within the current control cycle is calculated as the temperature fluctuation rate.

[0098] The dimensionless flow coefficient is obtained by obtaining the flow rate of the medium and dividing it by the rated flow rate reference.

[0099] The temperature deviation and the temperature fluctuation rate are respectively divided by the corresponding preset reference values ​​to perform dimensionless processing;

[0100] The product of the dimensionless flow coefficient and the dimensionless temperature deviation is used as the numerator, and the sum of the dimensionless temperature fluctuation rate and the preset constant is used as the denominator to calculate the ratio. The ratio is multiplied by the potential energy constant to obtain the original value of waste heat potential energy. The original value of waste heat potential energy is normalized according to the minimum and maximum original values ​​in the historical calibration samples, and the normalization result is limited to the range of 0 to 1 to obtain the waste heat potential energy factor.

[0101] Real-time data is acquired when calculating the waste heat potential energy factor. Assuming the current collected industrial waste heat source temperature is 45℃, a preset reference temperature of 25℃ is subtracted from this temperature to obtain a temperature deviation of 20℃. The absolute value of the time-varying rate of change of this industrial waste heat source temperature within the current 1-second control cycle is calculated. Assuming the temperature fluctuates at a rate of 0.5℃ / s, the temperature fluctuation rate is determined to be 0.5℃ / s. The flow meter measures the heat exchanger medium flow rate as 1.5 m³ / s. 3 / h, divide the flow rate by the system's calibrated rated flow rate benchmark of 1.0m. 3The dimensionless flow coefficient was calculated to be 1.5 per hour. Subsequently, the deviation and fluctuation data were dimensionlessly processed. The temperature deviation of 20℃ was divided by the preset reference value of 10℃ to obtain a dimensionless temperature deviation of 2.0, and the temperature fluctuation rate of 0.5℃ / s was divided by the reference value of 0.2℃ / s to obtain a dimensionless temperature fluctuation rate of 2.5. After dimensionless processing of the above parameters, the numerator was calculated by multiplying the dimensionless flow coefficient 1.5 by the dimensionless temperature deviation 2.0, and the denominator was calculated by adding the dimensionless temperature fluctuation rate 2.5 to the preset constant 1.0. The ratio of the two was approximately 0.857. This ratio of 0.857 was multiplied by the potential energy constant 0.8 to obtain the waste heat potential energy original factor of 0.685. Then, range normalization and 0-to-1 limiting processing were performed according to the minimum and maximum original factors of historical calibration samples to obtain the waste heat potential energy factor. If the historical calibration interval is consistent with the current scale, the normalized waste heat potential energy factor can be approximated as 0.685; if the original factor exceeds the range of 0 to 1, then... Perform range normalization and upper and lower limit truncation, where The original factor of waste heat potential energy. and These are the minimum and maximum values ​​from the historical calibration samples. This parameter serves as an input variable for subsequent model tuning, representing the current level of industrial waste heat and the potential for heat energy transfer to the refrigerant.

[0102] In one embodiment, the step of predicting the exhaust temperature for the next cycle using a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor includes:

[0103] Determine whether the value of the waste heat potential energy factor is not less than the model parameter switching threshold;

[0104] If the value is not less than the model parameter switching threshold, then the background value is constructed using an integral formula, and the development coefficient is set to the first preset development coefficient.

[0105] If the value is less than the model parameter switching threshold, the background value is constructed using a weighted average formula based on the three most recent data, and the development coefficient is set to the second preset development coefficient.

[0106] Wherein, the first preset development coefficient is less than the second preset development coefficient;

[0107] Using the first-order multivariate grey prediction model after adjusting the background value and the development coefficient, while keeping the development coefficient unchanged, only the grey action and the driving coefficients corresponding to each input variable are solved, and the exhaust temperature of the next cycle is output based on the solution results.

[0108] When using a first-order multivariate grey prediction model to predict the exhaust temperature for the next cycle, the controller evaluates the state of the waste heat potential energy factor. Based on the previously calculated waste heat potential energy factor value of 0.685, it is determined that this value is greater than the model parameter switching threshold of 0.6. At this time, the system uses an integral formula to construct the background value matrix of the grey model, and simultaneously sets the model's development coefficient to the first preset development coefficient. The integral formula for constructing the background value includes: performing linear interpolation on the cumulative generation sequence between adjacent sampling times, and calculating the integral average value within this interval as the background value; preferably, the background value is expressed as a weighted average or trapezoidal integral approximation of two adjacent cumulative values. This integral formula is equivalent to using a smoother background value to construct weights, in order to adapt to the working conditions with higher waste heat potential energy and more available heat. Conversely, if the waste heat potential energy factor calculated under another working condition is only 0.4, which is less than the threshold of 0.6, then a weighted average formula based on the three most recent historical data is used to construct the background value. For example, the most recent data is assigned a weight of 0.5, and adjacent historical data are assigned weights of 0.3 and 0.2 respectively, while the development coefficient is set to the second preset development coefficient. Preferably, the first preset development coefficient is 0.2, and the second preset development coefficient is 0.8. This weighted average formula is equivalent to increasing the influence of recent samples on the background value, making the model more sensitive to recent state changes when the waste heat potential energy is insufficient or fluctuates strongly. Development coefficients of 0.2 and 0.8 are both used as preset adjustment parameters input into the model, and the least squares method is only used to solve for the remaining model parameters. After completing the above model structure adjustment, the first-order multivariate grey prediction model with adjusted background value sequence and development coefficients is used to output the predicted exhaust temperature value for the next cycle, for example, 72.5℃. The dynamic change curve of the waste heat potential energy factor is illustrated as follows. Figure 2 As shown, this figure records the periodic fluctuation process of the waste heat potential energy factor over 100 seconds.

[0109] S3. Determine the control direction of the indoor temperature difference according to the current operating mode, and calculate the first frequency adjustment amount according to the patterned temperature difference between the target indoor temperature and the indoor temperature. Calculate the second frequency adjustment amount according to the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line. Calculate the frequency command by weighting and fusing the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor. When the waste heat potential energy factor is not less than the weighted fusion switching threshold, increase the weight of the second frequency adjustment amount. When the waste heat potential energy factor is lower than the weighted fusion switching threshold, increase the weight of the first frequency adjustment amount.

[0110] The target indoor temperature received by the microcontroller is compared with the actual indoor temperature, and the sign of the first temperature difference is determined according to the current operating mode. In heating mode, the target indoor temperature minus the actual indoor temperature is used as the first temperature difference; in cooling mode, the actual indoor temperature minus the target indoor temperature is used as the first temperature difference. The first temperature difference is input into the incremental proportional-integral-derivative (PID) algorithm module to calculate the first frequency adjustment amount used to meet the basic indoor heating and cooling load. Using the Newton-Raphson iterative algorithm, a point on the previously constructed target compression process trajectory line whose horizontal coordinate is equal to the currently collected exhaust pressure is found. The vertical coordinate of this point is extracted as the target exhaust temperature. The second temperature difference is obtained by subtracting the predicted next-cycle exhaust temperature from the target exhaust temperature. The second temperature difference is input into another positional PID algorithm module to calculate the second frequency adjustment amount used to maintain stable system operation. When the control cycle is short and the exhaust pressure change is less than a preset pressure change threshold, the current exhaust pressure is used to approximate the next-cycle exhaust pressure. When the exhaust pressure change is not less than the pressure change threshold, the next-cycle exhaust pressure is first predicted based on the historical pressure change rate, and the temperature corresponding to the predicted exhaust pressure on the target compression process trajectory line is used as the target exhaust temperature. An optimal classification boundary value is trained offline using the support vector machine algorithm combined with historical energy efficiency data. This value is then used as the weight fusion switching threshold and embedded in the microcontroller. The weighted fusion coefficients are calculated using a Logistic smooth transition function with the natural constant as the base.

[0111] When the calculated waste heat potential energy factor is greater than or equal to the weighted fusion switching threshold, it indicates that the industrial waste heat is sufficient and can be fully utilized. At this time, the output bias of the control Logistic smooth transition function is close to 1, thus assigning a larger weight value of more than 50% to the second frequency adjustment amount, and correspondingly reducing the weight value of the first frequency adjustment amount. When the waste heat potential energy factor is less than the weighted fusion switching threshold, it indicates that the waste heat quality is low, and the focus should be on meeting the indoor ambient temperature. At this time, the output bias of the Logistic smooth transition function is close to zero, assigning a larger weight value to the first frequency adjustment amount. The first frequency adjustment amount is multiplied by the corresponding weight and the second frequency adjustment amount is multiplied by the corresponding weight to obtain the weighted fusion frequency command. In this embodiment, the frequency command obtained by the weighted fusion is a comprehensive frequency increment. The controller adds the comprehensive frequency increment to the current operating frequency and limits it according to the minimum and maximum operating frequencies allowed by the compressor to obtain the final absolute operating frequency sent to the inverter. A control pulse sequence is generated using a space vector pulse width modulation algorithm and sent to the inverter to drive the HVAC inverter compressor to operate at the final absolute operating frequency.

[0112] In one embodiment, the step of weightedly fusing the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor to obtain a frequency command includes:

[0113] Determine whether the waste heat potential energy factor is not less than the weight fusion switching threshold;

[0114] When the weight is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is set to 0.8, and the weight of the first frequency adjustment amount is set to 0.2;

[0115] When the weight is less than the weighted fusion switching threshold, the weight of the first frequency adjustment is set to 0.8, and the weight of the second frequency adjustment is set to 0.2.

[0116] The first frequency adjustment amount and the second frequency adjustment amount are multiplied by their respective weights and then added together to obtain the comprehensive frequency increment. The frequency command is then generated based on the comprehensive frequency increment and the current operating frequency.

[0117] During the weighted fusion phase, the waste heat potential energy factor is invoked to determine whether it meets the preset weighted fusion switching threshold condition. Assuming the waste heat potential energy factor for the current cycle is 0.685, and confirming that this value is not less than the preset weighted fusion switching threshold of 0.5, the weight of the second frequency adjustment is set to 0.8, and the weight of the first frequency adjustment is set to 0.2. If the waste heat potential energy factor drops to 0.3, which is less than 0.5, the weight of the first frequency adjustment is set to 0.8, and the weight of the second frequency adjustment is set to 0.2. After determining the weight allocation, an addition operation is performed. For example, the calculated first frequency adjustment is +5Hz, and the second frequency adjustment is +10Hz. Combining the weights of 0.2 and 0.8, 9.0Hz is calculated. This 9.0Hz is a comprehensive frequency increment, not the absolute operating frequency directly sent to the inverter. If the current operating frequency is 45Hz, the controller adds 45Hz to 9.0Hz to obtain 54Hz, then limits it according to the preset frequency range before sending it to the inverter driver for execution. The variable frequency control weight switching characteristic is illustrated as follows: Figure 3 As shown.

[0118] In one embodiment, calculating the second frequency adjustment amount based on the deviation between the next cycle exhaust temperature and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line includes:

[0119] Find a target trajectory point on the target compression process trajectory line that corresponds to the current exhaust pressure with pressure coordinates, the target trajectory point including pressure coordinates and temperature coordinates;

[0120] Extract the temperature coordinates of the target trajectory points as the target exhaust temperature;

[0121] The exhaust temperature deviation is obtained by subtracting the next cycle exhaust temperature from the target exhaust temperature.

[0122] The basic adjustment amount is obtained by multiplying the exhaust temperature deviation by a preset proportional constant of 0.6.

[0123] The basic adjustment amount and the preset feedforward compensation amount are superimposed with a limited amplitude, and the direction of the superimposed second frequency adjustment amount is constrained to be consistent with the adjustment direction corresponding to the exhaust temperature deviation, so as to obtain the second frequency adjustment amount.

[0124] When calculating the second frequency adjustment, an interpolation point lookup is performed on the target compression process trajectory. Assuming the current exhaust pressure is 2.8 MPa, the curve node corresponding to 2.8 MPa on the pressure axis is located on the trajectory curve, and the ordinate temperature value of this node, for example, 59.8℃, is extracted as the target exhaust temperature. The predicted exhaust temperature for the next cycle is obtained, for example, 72.5℃. The target exhaust temperature of 59.8℃ is subtracted from the predicted temperature of 72.5℃ to calculate the second temperature difference as -12.7℃. After obtaining the second temperature difference, it is multiplied by a preset proportional constant of 0.6 to calculate the basic adjustment amount as -7.62Hz. This is combined with a preset feedforward compensation amount set by the system, and the feedforward compensation amount is limited so that its superposition does not change the sign direction of the basic adjustment amount; for example, when the basic adjustment amount is -7.62Hz, the feedforward compensation amount, after limiting, is +2.5Hz, resulting in -5.12Hz, still maintaining the trend of decreasing frequency adjustment. The inverter performs frequency adjustment based on the second frequency adjustment amount, bringing the thermodynamic cycle state closer to the target compression process trajectory. In this example, since the predicted exhaust temperature is higher than the target exhaust temperature, the exhaust temperature deviation is negative, and the second frequency adjustment amount also shows a trend of reducing the frequency, with its control direction consistent with the purpose of reducing the excessively high exhaust temperature.

[0125] Example 2

[0126] Embodiment 2 of the present invention proposes a variable frequency control system for HVAC that couples industrial waste heat, comprising the following modules:

[0127] The module is used to acquire indoor temperature, suction pressure, discharge pressure, suction temperature, discharge temperature, industrial waste heat source temperature and medium flow rate. Combined with the trajectory extension coefficient determined by the second frequency adjustment of the previous control cycle, the target compression process trajectory line spliced ​​by Bézier curves is constructed based on the suction pressure and the industrial waste heat source temperature. The control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature.

[0128] The adjustment module is used to calculate the waste heat potential energy factor based on the industrial waste heat source temperature, the heat source temperature change rate and the medium flow rate. Taking the intake state, the waste heat potential energy factor and the current operating frequency as input, the next cycle exhaust temperature is predicted by a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor.

[0129] The calculation module is used to determine the control direction of the indoor temperature difference according to the current operating mode, calculate the first frequency adjustment amount according to the patterned temperature difference between the target indoor temperature and the indoor temperature, calculate the second frequency adjustment amount according to the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line, and obtain a frequency command by weighted fusion of the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor. When the waste heat potential energy factor is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is increased; when the waste heat potential energy factor is lower than the weight fusion switching threshold, the weight of the first frequency adjustment amount is increased.

[0130] It should be understood that in the foregoing description of the embodiments in this specification, various features are combined in a single embodiment, drawing, or description for the purpose of simplifying the description and to aid in understanding a feature. However, this does not mean that the combination of these features is necessary, and those skilled in the art, upon reading this specification, may readily identify some of the devices as separate embodiments. That is, the embodiments in this specification can also be understood as an integration of multiple secondary embodiments. And the content of each secondary embodiment is valid even if it contains fewer than all the features of a single foregoing disclosed embodiment.

Claims

1. A variable frequency control method for HVAC systems coupled with industrial waste heat, characterized in that, Includes the following steps: The system acquires indoor temperature, suction pressure, discharge pressure, suction temperature, discharge temperature, industrial waste heat source temperature, and medium flow rate. Combined with the trajectory extension coefficient determined by the second frequency adjustment of the previous control cycle, a target compression process trajectory line spliced ​​with Bézier curves is constructed based on the suction pressure and the industrial waste heat source temperature. The control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature. The waste heat potential energy factor is calculated based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the medium flow rate. The exhaust temperature of the next cycle is predicted by a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor, with the intake state, the waste heat potential energy factor, and the current operating frequency as inputs. The control direction of the indoor temperature difference is determined according to the current operating mode, and the first frequency adjustment amount is calculated according to the patterned temperature difference between the target indoor temperature and the indoor temperature. The second frequency adjustment amount is calculated according to the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line. The first frequency adjustment amount and the second frequency adjustment amount are weighted and fused based on the waste heat potential energy factor to obtain the frequency command. When the waste heat potential energy factor is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is increased. When the waste heat potential energy factor is lower than the weight fusion switching threshold, the weight of the first frequency adjustment amount is increased.

2. The method according to claim 1, characterized in that, The step of constructing a target compression process trajectory line by splicing Bezier curves based on the suction pressure and the industrial waste heat source temperature, wherein the control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature, includes: Obtain the preset maximum upper limit exhaust pressure, and subtract the current intake pressure to get the pressure range difference; The pressure interval difference is divided into three parts, and the pressure coordinates of the starting endpoint, two segment splicing nodes and the ending endpoint are determined. The pressure coordinates of two internal control points are generated between each adjacent endpoint according to a preset ratio, forming the segment endpoint and internal control point coordinate reference required for three third-order Bézier curves. The refrigerant suction state entropy value is determined based on the current suction pressure and current suction temperature, and the corresponding isentropic compression theoretical temperature is calculated based on the suction state entropy value and each pressure coordinate reference, which serves as the base temperature for the corresponding control point. Obtain the refrigerant saturation temperature corresponding to the current suction pressure, and calculate the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source. The temperature coordinate offset of the control point is obtained by multiplying the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source by a proportional constant and combining it with the trajectory extension coefficient. The temperature coordinates are obtained by adding the base temperature corresponding to each pressure coordinate reference with the temperature coordinate offset. The pressure coordinates of the segment endpoints, splicing nodes and internal control points are combined with the corresponding temperature coordinates to generate the control point coordinates of three third-order Bézier curves. The three Bézier curves are generated by the control point coordinates and smoothly spliced ​​to output the target compression process trajectory line.

3. The method according to claim 2, characterized in that, The calculation of the waste heat potential energy factor based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the flow rate of the medium includes: Calculate the temperature deviation obtained by subtracting the preset reference temperature from the current temperature of the industrial waste heat source, and perform non-negative truncation processing on the temperature deviation. The absolute value of the time-varying rate of temperature change of the industrial waste heat source within the current control cycle is calculated as the temperature fluctuation rate. The dimensionless flow coefficient is obtained by obtaining the flow rate of the medium and dividing it by the rated flow rate reference. The temperature deviation and the temperature fluctuation rate are respectively divided by the corresponding preset reference values ​​to perform dimensionless processing; The product of the dimensionless flow coefficient and the dimensionless temperature deviation is used as the numerator, and the sum of the dimensionless temperature fluctuation rate and the preset constant is used as the denominator to calculate the ratio. The ratio is multiplied by the potential energy constant to obtain the original value of waste heat potential energy. The original value of waste heat potential energy is normalized according to the minimum and maximum original values ​​in the historical calibration samples, and the normalization result is limited to the range of 0 to 1 to obtain the waste heat potential energy factor.

4. The method according to claim 1, characterized in that, The prediction of the next cycle exhaust temperature using a first-order multivariate grey prediction model with parameters adjusted by the waste heat potential energy factor includes: Determine whether the value of the waste heat potential energy factor is not less than the model parameter switching threshold; If the value is not less than the model parameter switching threshold, then the background value is constructed using an integral formula, and the development coefficient is set to the first preset development coefficient. If the value is less than the model parameter switching threshold, the background value is constructed using a weighted average formula based on the three most recent data, and the development coefficient is set to the second preset development coefficient. Wherein, the first preset development coefficient is less than the second preset development coefficient; Using the first-order multivariate grey prediction model after adjusting the background value and the development coefficient, while keeping the development coefficient unchanged, only the grey action and the driving coefficients corresponding to each input variable are solved, and the exhaust temperature of the next cycle is output based on the solution results.

5. The method according to claim 1 or 4, characterized in that, The step of weighting and fusing the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor to obtain the frequency command includes: Determine whether the waste heat potential energy factor is not less than the weight fusion switching threshold; When the weight is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is set to 0.8, and the weight of the first frequency adjustment amount is set to 0.2; When the weight is less than the weighted fusion switching threshold, the weight of the first frequency adjustment is set to 0.8, and the weight of the second frequency adjustment is set to 0.

2. The first frequency adjustment amount and the second frequency adjustment amount are multiplied by their respective weights and then added together to obtain the comprehensive frequency increment. The frequency command is then generated based on the comprehensive frequency increment and the current operating frequency.

6. The method according to claim 1, characterized in that, The step of calculating the second frequency adjustment amount based on the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line includes: Find a target trajectory point on the target compression process trajectory line that corresponds to the current exhaust pressure with pressure coordinates, the target trajectory point including pressure coordinates and temperature coordinates; Extract the temperature coordinates of the target trajectory points as the target exhaust temperature; The exhaust temperature deviation is obtained by subtracting the next cycle exhaust temperature from the target exhaust temperature. The basic adjustment amount is obtained by multiplying the exhaust temperature deviation by a preset proportional constant of 0.

6. The basic adjustment amount and the preset feedforward compensation amount are superimposed with a limited amplitude, and the direction of the superimposed second frequency adjustment amount is constrained to be consistent with the adjustment direction corresponding to the exhaust temperature deviation, so as to obtain the second frequency adjustment amount.

7. A variable frequency control system for HVAC coupled with industrial waste heat, characterized in that, Includes the following modules: The module is used to acquire indoor temperature, suction pressure, discharge pressure, suction temperature, discharge temperature, industrial waste heat source temperature and medium flow rate. Combined with the trajectory extension coefficient determined by the second frequency adjustment of the previous control cycle, the target compression process trajectory line spliced ​​by Bézier curves is constructed based on the suction pressure and the industrial waste heat source temperature. The control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature. The adjustment module is used to calculate the waste heat potential energy factor based on the industrial waste heat source temperature, the heat source temperature change rate and the medium flow rate. Taking the intake state, the waste heat potential energy factor and the current operating frequency as input, the next cycle exhaust temperature is predicted by a first-order multivariate grey prediction model whose parameters are adjusted by the waste heat potential energy factor. The calculation module is used to determine the control direction of the indoor temperature difference according to the current operating mode, calculate the first frequency adjustment amount according to the patterned temperature difference between the target indoor temperature and the indoor temperature, calculate the second frequency adjustment amount according to the deviation between the exhaust temperature of the next cycle and the target temperature corresponding to the exhaust pressure on the target compression process trajectory line, and obtain a frequency command by weighted fusion of the first frequency adjustment amount and the second frequency adjustment amount based on the waste heat potential energy factor. When the waste heat potential energy factor is not less than the weight fusion switching threshold, the weight of the second frequency adjustment amount is increased; when the waste heat potential energy factor is lower than the weight fusion switching threshold, the weight of the first frequency adjustment amount is increased.

8. The system according to claim 7, characterized in that, The step of constructing a target compression process trajectory line by splicing Bezier curves based on the suction pressure and the industrial waste heat source temperature, wherein the control points of the target compression process trajectory line are mapped according to the difference between the refrigerant saturation temperature corresponding to the suction pressure and the industrial waste heat source temperature, includes: Obtain the preset maximum upper limit exhaust pressure, and subtract the current intake pressure to get the pressure range difference; The pressure interval difference is divided into three parts, and the pressure coordinates of the starting endpoint, two segment splicing nodes and the ending endpoint are determined. The pressure coordinates of two internal control points are generated between each adjacent endpoint according to a preset ratio, forming the segment endpoint and internal control point coordinate reference required for three third-order Bézier curves. The refrigerant suction state entropy value is determined based on the current suction pressure and current suction temperature, and the corresponding isentropic compression theoretical temperature is calculated based on the suction state entropy value and each pressure coordinate reference, which serves as the base temperature for the corresponding control point. Obtain the refrigerant saturation temperature corresponding to the current suction pressure, and calculate the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source. The temperature coordinate offset of the control point is obtained by multiplying the difference between the refrigerant saturation temperature corresponding to the current suction pressure and the temperature of the industrial waste heat source by a proportional constant and combining it with the trajectory extension coefficient. The temperature coordinates are obtained by adding the base temperature corresponding to each pressure coordinate reference with the temperature coordinate offset. The pressure coordinates of the segment endpoints, splicing nodes and internal control points are combined with the corresponding temperature coordinates to generate the control point coordinates of three third-order Bézier curves. The three Bézier curves are generated by the control point coordinates and smoothly spliced ​​to output the target compression process trajectory line.

9. The system according to claim 7, characterized in that, The calculation of the waste heat potential energy factor based on the industrial waste heat source temperature, the rate of change of the heat source temperature, and the flow rate of the medium includes: Calculate the temperature deviation obtained by subtracting the preset reference temperature from the current temperature of the industrial waste heat source, and perform non-negative truncation processing on the temperature deviation. The absolute value of the time-varying rate of temperature change of the industrial waste heat source within the current control cycle is calculated as the temperature fluctuation rate. The dimensionless flow coefficient is obtained by obtaining the flow rate of the medium and dividing it by the rated flow rate reference. The temperature deviation and the temperature fluctuation rate are respectively divided by the corresponding preset reference values ​​to perform dimensionless processing; The product of the dimensionless flow coefficient and the dimensionless temperature deviation is used as the numerator, and the sum of the dimensionless temperature fluctuation rate and the preset constant is used as the denominator to calculate the ratio. The ratio is multiplied by the potential energy constant to obtain the original value of waste heat potential energy. The original value of waste heat potential energy is normalized according to the minimum and maximum original values ​​in the historical calibration samples, and the normalization result is limited to the range of 0 to 1 to obtain the waste heat potential energy factor.

10. The system according to claim 7, characterized in that, The prediction of the next cycle exhaust temperature using a first-order multivariate grey prediction model with parameters adjusted by the waste heat potential energy factor includes: Determine whether the value of the waste heat potential energy factor is not less than the model parameter switching threshold; If the value is not less than the model parameter switching threshold, then the background value is constructed using an integral formula, and the development coefficient is set to the first preset development coefficient. If the value is less than the model parameter switching threshold, the background value is constructed using a weighted average formula based on the three most recent data, and the development coefficient is set to the second preset development coefficient. Wherein, the first preset development coefficient is less than the second preset development coefficient; Using the first-order multivariate grey prediction model after adjusting the background value and the development coefficient, while keeping the development coefficient unchanged, only the grey action and the driving coefficients corresponding to each input variable are solved, and the exhaust temperature of the next cycle is output based on the solution results.

Images