Surge early warning method and device for gas compressor
By collecting and processing signals during compressor operation, calculating entropy values and outputting early warning signals, the performance degradation and safety risks caused by compressor surge are solved, and the monitoring and early warning of surge conditions are achieved.
Patent Information
- Application Number
- CN202411251282.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-07
- Publication Date
- 2024-12-03
AI Technical Summary
Instability and surge phenomena caused by intake distortion during operation of the compressor, resulting in performance degradation and may even cause structural damage and accidents. Effective monitoring and early warning mechanisms are needed to avoid these problems.
By collecting dynamic pressure signals, sound signals and vibration signals during the operation of the compressor, calculate their entropy values, and performing weighting processing. If the preset change rules are met, an early warning signal will be output to monitor and warning the surge state.
The monitoring and early warning of the compressor surge status is realized, the surge prediction efficiency is improved, and the performance and safety problems caused by surge are reduced.
Smart Images

Figure CN119062603A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of compressor surge warning, and in particular to a compressor surge warning method and device. Background Art
[0002] During the operation of the compressor, the instability and surge of the compressor system caused by reasons such as intake distortion is one of the main problems that restrict the performance of the compressor. The occurrence of surge will cause the performance of the compressor to decline rapidly, resulting in a sharp deterioration of the compressor characteristics. In severe cases, it will cause the compressor to shut down, or even structural damage such as blade breakage, causing serious accidents and losses. Therefore, during the use of the compressor, in order to ensure its normal and stable operation, it is necessary to monitor the operating status of the compressor to prevent it from entering an unstable or even surge state. Summary of the invention
[0003] In view of this, the embodiments of the present application hope to provide a compressor surge warning method and device to at least solve the above-mentioned technical problems.
[0004] To achieve the above purpose, the technical solution of this application is implemented as follows:
[0005] According to one aspect of an embodiment of the present application, a compressor surge warning method is provided, the method comprising:
[0006] Collecting a first target signal generated by the compressor during operation, wherein the first target signal represents an operating state of the compressor;
[0007] Determining entropy values corresponding to the first target signal under different operating conditions;
[0008] Performing weighted processing on the entropy value to obtain a weighted entropy value;
[0009] If the weighted entropy value and / or the change state of the entropy value meets a preset change rule, a warning signal is output, and the warning signal represents the surge state of the compressor.
[0010] In the above solution, the first target signal generated by the compressor during operation is collected, including:
[0011] collecting a first target signal generated by the compressor during operation at intervals according to preset time parameters based on a preset time window, wherein the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal;
[0012] Or, a first target signal generated by the compressor during operation is continuously collected based on a preset time window, and the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal.
[0013] In the above solution, the weighted processing of the entropy value to obtain the weighted entropy value includes:
[0014] Processing the entropy value to obtain an information entropy value of the entropy value;
[0015] Determine a collection end weight of the first target signal based on the information entropy value;
[0016] A weighted entropy value of the entropy value is determined based on the acquisition end weight.
[0017] In the above solution, determining the entropy value corresponding to the first target signal under different operating conditions includes:
[0018] Determine at least one of an information entropy value, a sample entropy value, a fuzzy entropy value, and an approximate entropy value corresponding to the first target signal under different operating conditions.
[0019] In the above solution, the first target signal is a time series signal under any operating condition; determining the information entropy value corresponding to the first target signal under different operating conditions includes:
[0020] The first target signal is recorded as X, which is expressed as:
[0021] X=[u(1),u(2),u(3),…,u(N)];
[0022] Wherein, u(1) represents the value of the first target signal collected at the first sampling moment, u(2) represents the value of the first target signal collected at the second moment, u(3) represents the value of the first target signal collected at the third moment, and so on, u(N) represents the value of the first target signal collected at the Nth moment;
[0023] Among them, the value of the first target signal has multiple results, and the probability of each result is p(x), which is expressed as:
[0024]
[0025] The information entropy value corresponding to the first target signal is recorded as InEn(X) and is calculated as follows:
[0026] InEn(X)=-∑p(x)log2 p(x);
[0027] Wherein, num(u=x) represents the number of first target signal values x, and N represents the number of all first target signal values;
[0028] The first target signal is a time series signal under any operating condition; determining the sample entropy value corresponding to the first target signal under different operating conditions includes:
[0029] The first target signal is recorded as X, which is expressed as:
[0030] X=[u(1),u(2),u(3),…,u(N)];
[0031] Wherein, u(1) represents the value of the first target signal collected at the first sampling moment, u(2) represents the value of the first target signal collected at the second moment, u(3) represents the value of the first target signal collected at the third moment, and so on, u(N) represents the value of the first target signal collected at the Nth moment;
[0032] Any subsequence of the first target signal starting from the i-th sampling time is denoted as X(i) and is expressed as:
[0033] X(i)=[u(i),u(i+1),…,u(i+m-1)];
[0034] Wherein, i represents the i-th sampling moment, i is any integer between 1 and N-m+1; x(i) is the value of the first target signal collected at the i-th sampling moment; m is the preset sequence dimension value, ranging from 1 to Ni;
[0035] Under the condition of determining the value of i, traverse all combinations of X(i) and X(j) and calculate the maximum distance between X(i) and X(j), which is recorded as: d[X(i), X(j)], and the calculation formula is:
[0036] d[X(i),X(j)]=max k=0,...,m-1 |u(i+k)-u(j+k)|;
[0037] Wherein, X(i) represents any subsequence of the first target signal starting from the i-th sampling moment; X(j) represents any subsequence of the first target signal starting from the j-th sampling moment, i is not equal to, i and j are any integers j between 1 and N-m+1; k is any integer between 0 and m-1, max represents the maximum value of all |u(i+k)-u(j+k)| calculated from k from 0 to m-1; u(i+k) and u(j+k) represent the values of the first target signal collected at the i+k and j+k sampling moments, respectively;
[0038] Based on the distance between X(i) and X(j), we count the number of items between X(i) and X(j) that meet the requirements in 1≤i≤Nm and 1≤j≤Nm and j≠i, which is denoted as B i (r), the calculation formula is:
[0039] B i (r)=num{d[X(i),X(j)]≤r};
[0040] The condition requires that the distance between X(i) and X(j) (d[X(i), X(j)]) is less than or equal to the preset parameter r value, and each d[X(i), X(j)] is regarded as 1 item, and the items that meet the condition are recorded as 1, and the items that do not meet the condition are recorded as 0;
[0041] Find the ratio of the number of items that meet the conditional requirements to the total number, denoted as The calculation formula is:
[0042]
[0043] Then find the logarithmic mean of the ratio, denoted as Φ m (r), the calculation formula is:
[0044]
[0045] Among them, Nm means that under the condition of determining the value of i, all combinations of X(i) and X(j) are traversed, and there are Nm combinations, that is, the total number;
[0046] The sample entropy value corresponding to the first target signal is denoted as SaEn, and the calculation formula is:
[0047]
[0048] Among them, the sequence dimension m is added by 1 to get m+1, and Φ m (r) and Φ m+1 (r) is obtained in the same way.
[0049] In the above scheme, if the weighted entropy value and / or the change state of the entropy value satisfies a preset change rule, outputting a warning signal includes:
[0050] Constructing a curve model based on the weighted entropy value and / or the entropy value corresponding to the different operating conditions;
[0051] If the curve model presents at least one of a W-type and an N-type structure, it is determined that the weighted entropy value and / or the change state of the entropy value satisfies a preset change rule, and a warning signal is output.
[0052] In the above scheme, the warning levels of the warning signals corresponding to the W type and the N type are different; wherein the warning level corresponding to the W type is higher than the warning level corresponding to the N type.
[0053] In the above scheme, the method further comprises:
[0054] comparing the weighted entropy value and / or the entropy value with a preset surge probability value;
[0055] If the comparison result indicates that the weighted entropy value and / or the difference between the entropy value and the surge probability value is less than a first threshold, outputting a first warning signal, wherein the first warning signal indicates that the surge state of the compressor is close to a surge boundary;
[0056] If the comparison result indicates that the weighted entropy value and / or the entropy value is greater than or equal to the surge probability value, a second warning signal is output, wherein the second warning signal indicates that the surge state of the compressor is that surge has occurred.
[0057] In the above solution, the weighted entropy value and / or the entropy value changes with the change of air parameters in the compressor.
[0058] According to another aspect of the present application, a surge warning device for a compressor is provided, the device comprising:
[0059] A collection unit, used for collecting a first target signal generated by the compressor during operation, wherein the first target signal represents an operating state of the compressor;
[0060] A determination unit, configured to determine an entropy value corresponding to the first target signal under different operating conditions;
[0061] A processing unit, used for performing weighted processing on the entropy value to obtain a weighted entropy value;
[0062] The output unit is used to output a warning signal if the weighted entropy value and / or the change state of the entropy value meets a preset change rule, and the warning signal represents the surge state of the compressor.
[0063] According to a third aspect of the present application, a surge warning device for a compressor is provided, the device comprising: a processor and a memory for storing a computer program that can be run on the processor, the processor being configured to execute any of the surge warning methods described above when running the computer program.
[0064] The surge warning method and device for a compressor provided in the present application is a method for monitoring the surge probability of the compressor during operation by means of the weighted entropy value of the compressor. Specifically, by collecting the first target signal generated by the compressor during operation, the first target signal represents the operating state of the compressor; determining the entropy value corresponding to the first target signal under different operating conditions; performing weighted processing on the entropy value to obtain a weighted entropy value; if the weighted entropy value and / or the change state of the entropy value meets the preset change rule, outputting a warning signal, the warning signal represents the surge state of the compressor. In this way, not only can the surge state of the compressor be monitored, but also different surge states of the compressor can be warned in advance, which greatly improves the surge prediction efficiency of the compressor and reduces the performance safety problems of the compressor caused by surge. BRIEF DESCRIPTION OF THE DRAWINGS
[0065] Figure 1 The flow diagram of the compressor surge warning method in this application is as follows Figure 1 ;
[0066] Figure 2 The flow diagram of the compressor surge warning method in this application is as follows Figure 2 ;
[0067] Figure 3 It is a curve diagram of the weighted information entropy value of the aerodynamic noise time series signal at the intake end of the axial compressor;
[0068] Figure 4 The schematic diagram of the structure of the compressor surge warning device in this application is as follows Figure 1 ;
[0069] Figure 5 The schematic diagram of the structure of the compressor surge warning device in this application is as follows Figure 2 . DETAILED DESCRIPTION
[0070] The technical solution of the present application is further elaborated in detail below in conjunction with the accompanying drawings and specific embodiments of the specification.
[0071] The various specific technical features in the various embodiments described in the specific implementation methods can be combined in various ways without contradiction. For example, different implementation methods can be formed by combining different specific technical features. In order to avoid unnecessary repetition, the various possible combinations of the specific technical features in this application will not be described separately.
[0072] It should be noted that the terms "first\second\third" involved in the embodiments of the present application are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that the specific order or sequence of "first\second\third" can be interchanged where permitted. It should be understood that the objects distinguished by "first\second\third" can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.
[0073] Figure 1 The schematic diagram of the process of implementing the surge warning method of the compressor in this application is as follows Figure 1 ,This method can be applied to the surge warning device of the compressor, such as Figure 1 As shown, the method includes:
[0074] Step 101, collecting a first target signal generated by a compressor during operation, wherein the first target signal represents an operating state of the compressor;
[0075] Here, the first target signal includes but is not limited to a sound signal, a pressure signal, and a vibration signal generated by the compressor during operation.
[0076] The pressure signal may be a transient pressure signal of the internal flow field of the inner wall of the wheel rim casing of the compressor, or a pressure signal of a certain spatial point in the flow field.
[0077] The sound signal can be the pressure field noise inside the compressor or the free field noise outside the compressor.
[0078] The vibration signal can come from the structural vibration of the compressor itself or from the shaft vibration of the rotating part of the compressor.
[0079] Step 102, determining the entropy value corresponding to the first target signal under different operating conditions;
[0080] Here, the entropy value includes but is not limited to an approximate entropy value, a fuzzy entropy value, an information entropy value, and a sample entropy value.
[0081] Step 103, performing weighted processing on the entropy value to obtain a weighted entropy value;
[0082] Here, the weighted entropy value includes but is not limited to an approximate entropy weighted value, a fuzzy entropy weighted value, an information entropy weighted value, and a sample entropy weighted value.
[0083] Step 104: if the weighted entropy value and / or the change state of the entropy value satisfies a preset change rule, output a warning signal, wherein the warning signal indicates the surge state of the compressor.
[0084] Here, the early warning device can construct a curve model based on the weighted entropy value and / or entropy value corresponding to the compressor under different operating conditions; if the curve model is in at least one of a W-type and an N-type structure, it is determined that the change state of the weighted entropy value and / or the entropy value meets the preset change law, and an early warning signal is output. If the curve model is in a V-type structure, it indicates that the change state of the weighted entropy value and / or the entropy value does not meet the preset change law, and no early warning signal is output.
[0085] Here, the warning levels of the warning signals corresponding to the W type and the N type are different; wherein the warning level corresponding to the W type is higher than the warning level corresponding to the N type.
[0086] for example Figure 2As shown in the figure, the early warning device can monitor the first target signal generated during the operation of the compressor through the pressure sensor 201 and the sound sensor 202. The first target signal can be a time series signal, and the entropy value of the first target signal under different operating conditions (including but not limited to sample entropy, approximate entropy, fuzzy entropy and / or information entropy) is calculated, and the flow rate of the compressor can be monitored through the flow meter 203. After calculating the entropy values under different operating conditions under multiple sensors of the same type, the weighted entropy values of the same entropy value under different operating conditions can also be calculated. By tracking the change law of the weighted entropy value corresponding to the compressor under different operating conditions as the compressor gradually decreases from the near-blockage flow; when the weighted entropy value shows a downward trend for the first time as the compressor flow decreases, it is considered that the compressor is running stably (normally) and no early warning will be issued; when the weighted entropy value shows an upward trend for the first time as the compressor flow decreases, it is considered that the compressor is running stably (normally) and no early warning will be issued; when the weighted entropy value decreases with the compressor flow and shows a downward trend for the second time after the first downward trend, it is determined that the change state of the weighted entropy value meets A preset change law is followed, and a primary warning signal is output, which indicates that the operating condition of the compressor is close to the surge boundary. When the entropy value decreases with the compressor flow and presents an upward trend for the second time, it is determined that the change state of the weighted entropy value meets the preset change law, and an intermediate warning signal is output, which indicates that the operating condition of the compressor is further close to the surge boundary. When the weighted entropy value decreases with the compressor flow and suddenly jumps after presenting an upward trend for the second time, an advanced warning signal is output, which indicates that surge has occurred in the compressor.
[0087] Here, the warning methods of the primary warning signal, the intermediate warning signal and the advanced warning signal are not limited, and can be one or a combination of voice warning, light warning, vibration warning, and sound warning.
[0088] The compressor surge warning method provided in the present application can realize surge prediction under different operating conditions of the compressor by monitoring the entropy value and weighted entropy value corresponding to the first target signal generated during the operation of the compressor under different operating conditions, thereby enabling early warning of the surge probability, which can greatly improve the safety performance during the operation of the compressor and avoid serious accidents and losses.
[0089] In the present application, when the early warning device collects the first target signal, it is not necessary to limit the stability of the compressor operating condition.
[0090] For example, the early warning device can collect the first target signal generated by the compressor during the stable operation of a certain operating point of the compressor. The stability of a certain operating point of the compressor can indicate that the compressor flow, pressure ratio and efficiency remain unchanged or change periodically.
[0091] For another example, the early warning device may also collect the first target signal generated by the compressor during the operation process when the operating point of the compressor changes. The change in the operating point of the compressor may indicate that the compressor changes from a certain stable operating point to another operating point different from the previous operating state.
[0092] In the present application, the collection method of the early warning device when collecting the first target signal may not be limited.
[0093] For example, the early warning device may collect a first target signal generated by the compressor during operation at intervals according to a preset time parameter based on a preset time window, wherein the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal;
[0094] Alternatively, the early warning device may also continuously collect a first target signal generated by the compressor during operation based on a preset time window, wherein the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal.
[0095] In the present application, the first target signal may be a time series signal under any operating condition; the early warning device may determine the approximate entropy value corresponding to the first target signal under different operating conditions, and the calculation method of the approximate entropy value is as follows:
[0096] For example, the first target signal is denoted as X, which can be regarded as a vector and expressed as:
[0097] X=[x(1),x(2),x(3),…,x(N)];
[0098] Among them, x(1) represents the first target signal value collected at the first sampling moment, x(2) represents the first target signal value collected at the second moment, x(3) represents the first target signal value collected at the third moment, and so on, x(N) represents the first target signal value collected at the Nth moment; N can be the number of sampling time points or the dimension of the first target signal.
[0099] Any subsequence of the first target signal starting from the i-th sampling time, denoted as X(i), can also be regarded as a subvector, expressed as:
[0100] X(i)=[x(i),x(i+1),…,x(i+m-1)];
[0101] Among them, i is any integer between 1 and N-m+1; m is a preset sequence dimension, ranging from 1 to Ni.
[0102] Under the condition of determining the value of i, traverse all combinations of sub-vectors X(i) and X(j), and calculate the maximum distance between sub-vectors X(i) and X(j), which is recorded as: d[X(i), X(j)], and the calculation formula is:
[0103] d[X(i),X(j)]=max k=0,...,m-1 |u(i+k)-u(j+k)|;
[0104] Wherein, X(i) represents any subsequence of the first target signal starting from the i-th sampling moment; X(j) represents any subsequence of the first target signal starting from the j-th sampling moment, and i may be equal to or not equal to j;
[0105] Based on the distance between X(i) and X(j), the number of items that meet the conditional requirements is calculated, denoted as B i (r), also known as the approximate number, is calculated as:
[0106] B i (r) = num{d[X i , X j ]<r};
[0107] The condition requirement may be that the distance d[X(i), X(j)] between X(i) and X(j) is less than a preset parameter r value; each d[X(i), X(j)] is regarded as 1 item, and the item that meets the condition requirement is recorded as 1, and the item that does not meet the condition requirement is recorded as 0;
[0108] Find the ratio of the number of items (or approximate number) that meet the conditions to the total number, recorded as The calculation formula is:
[0109]
[0110] Then find the logarithmic mean of the ratio, denoted as Φ m (r), the calculation formula is:
[0111]
[0112] Among them, N-m+1 means that under the condition of determining the value of i, traversing all the combinations of X(i) and X(j), there are N-m+1 combinations, that is, the total number;
[0113] The approximate entropy value corresponding to the first target signal is denoted as f ApEn , the calculation formula is:
[0114] f ApEn =Φ m (r)-Φ m+1 (r);
[0115] In which, the sequence dimension m is added by 1 to get m+1, and and Φ m That is, by repeating the above steps, we can obtain
[0116] Here, approximate entropy is a concept derived from statistics, which is used to quantitatively characterize the probability of generating new patterns in signals. The larger the value, the more complex the time series signal is, and the stronger its disorder and unpredictability are. Approximate entropy has good anti-noise and anti-interference capabilities, and the analysis effect is better than conventional statistical methods such as variance and standard deviation. It also requires fewer data points and has good versatility. Therefore, by monitoring the surge probability of the compressor through the method of approximate entropy, surge warning can be carried out in advance, which greatly improves the safety performance of the compressor and avoids serious losses caused by surge of the compressor.
[0117] In the present application, the first target signal may be a time series signal under any operating condition; the early warning device may also determine the fuzzy entropy value corresponding to the first target signal under different operating conditions, and the calculation method of the fuzzy entropy value is as follows:
[0118] For example, the first target signal is recorded as X, which can also be called a vector and is expressed as:
[0119] X=[x(1),x(2),x(3),…,x(N)];
[0120] Wherein, x(1) represents the first target signal value collected at the first sampling moment, x(2) represents the first target signal value collected at the second moment, x(3) represents the first target signal value collected at the third moment, and so on, x(N) represents the first target signal value collected at the Nth moment;
[0121] Any subsequence of the first target signal starting from the i-th sampling time is denoted as X(i), which can also be called a subvector and is expressed as:
[0122] X(i)=[x(i),x(i+1),…,x(i+m-1)]-x0(i);
[0123] Where i represents the i-th sampling moment, X(i) is the first target signal value collected at the i-th sampling moment; i is any integer between 1 and N-m+1; m is the preset sequence dimension value, ranging from 1 to Ni; x0(i) represents the mean of m consecutive x(i), and the calculation formula is:
[0124]
[0125] Under the condition of determining the value of i, traverse all combinations of sub-vectors X(i) and X(j), and calculate the distance between sub-vectors X(i) and X(j), which is recorded as: d[X(i), X(j)], and the calculation formula is:
[0126] d[X(i),X(j)]=max k=0,...,m-1 ||x(i+k)-x0(i)|-|x(j+k)-x0(j)||;
[0127] Wherein, X(i) represents any subsequence of the first target signal starting from the i-th sampling time; X(j) represents any subsequence of the first target signal starting from the j-th sampling time, and i is not equal to j;
[0128] On the basis of obtaining the distance between sub-vectors X(i) and X(j), the number of sub-vectors X(i) and X(j) whose distance (d[X(i), X(j)]) is less than the preset parameter r is calculated, which is recorded as The calculation formula is:
[0129]
[0130] Where n is a preset value, and e is the base of the natural logarithm;
[0131] Ask for it again The average of m (n, r), the calculation formula is:
[0132]
[0133] Among them, each d[X(i), X(j)] is regarded as 1 item, there will be Nm items, that is, the total number is Nm;
[0134] The fuzzy entropy value corresponding to the first target signal is recorded as FuzzyEn(m, n, r), and the calculation formula is:
[0135] FuzzyEn(m,n,r)=lnφ m (n, r)-lnφ m+1 (n, r);
[0136] Among them, the sequence dimension m is increased by 1 to get m+1, and φ m+1 (n, r) and φ m (n, r) is obtained in the same way, that is, by repeating the above formula steps, φ can be obtained m+1 (n, r).
[0137] In the present application, the first target signal may be a time series signal under any operating condition; the early warning device may also determine the sample entropy value corresponding to the first target signal under different operating conditions, and the calculation method of the sample entropy value is as follows:
[0138] For example, the first target signal is recorded as X, which is expressed as:
[0139] X=[u(1),u(2),u(3),…,u(N)];
[0140] Among them, u(1) represents the first target signal value collected at the first sampling moment, u(2) represents the first target signal value collected at the second moment, u(3) represents the first target signal value collected at the third moment, and so on, u(N) represents the first target signal value collected at the Nth moment; N can be the number of sampling time points or the dimension of the first target signal.
[0141] Any subsequence of the first target signal starting from the i-th sampling time, denoted as X(i), can also be regarded as a subvector, expressed as:
[0142] X(i)=[u(i),u(i+1),…,u(i+m-1)] ;
[0143] Wherein, i represents the i-th sampling moment, i is any integer between 1 and N-m+1; u(i) is the first target signal value collected at the i-th sampling moment; m is the preset sequence dimension value, ranging from 1 to Ni;
[0144] Under the condition of determining the value of i, traverse all combinations of X(i) and X(j) and calculate the maximum distance between X(i) and X(j), which is recorded as: d[X(i), X(j)], and the calculation formula is:
[0145] d[X(i),X(j)]=max k=0,....m-1 |u(i+k)-u(j+k)|;
[0146] Wherein, X(i) represents any subsequence of the first target signal starting from the i-th sampling moment; X(j) represents any subsequence of the first target signal starting from the j-th sampling moment, i is not equal to j, i and j are any integers j between 1 and N-m+1; u(i+k) and u(j+k) represent the values of the first target signal collected at the i+k-th sampling moment and the j+k-th sampling moment, respectively; k is any integer between 0 and m-1, and max represents the maximum value of all |u(i+k)-u(j+k)| calculated from k from 0 to m-1;
[0147] Based on the distance between X(i) and X(j), we count the number of items between X(i) and X(j) that meet the requirements in 1≤i≤Nm and 1≤j≤Nm and j≠i, which is denoted as B i (r), the calculation formula is:
[0148] B i (r)=num{d[X(i),X(j)]≤r};
[0149] The condition requires that the distance between X(i) and X(j) (d[X(i), X(j)]) is less than or equal to the preset parameter r value, and each d[X(i), X(j)] is regarded as 1 item, and the items that meet the condition are recorded as 1, and the items that do not meet the condition are recorded as 0;
[0150] Find the ratio of the number of items that meet the conditional requirements to the total number, denoted as The calculation formula is:
[0151]
[0152] Then find the logarithmic mean of the ratio, denoted as Φ m (r), the calculation formula is:
[0153]
[0154] Among them, Nm means that under the condition of determining the value of i, all combinations of X(i) and X(j) are traversed, and there are Nm combinations, that is, the total number;
[0155] The sample entropy value corresponding to the first target signal is denoted as SaEn, and the calculation formula is:
[0156]
[0157] Among them, the sequence dimension m is added by 1 to get m+1, and Φ m (r) and Φ m+1 (r) is obtained in the same way.
[0158] In the present application, the first target signal may be a time series signal under any operating condition; the early warning device may also determine the information entropy value corresponding to the first target signal under different operating conditions, and the calculation method of the information entropy value is as follows:
[0159] For example, the first target signal is denoted as X, which is expressed as:
[0160] X=[u(1),u(2),u(3),…,u(N)];
[0161] Wherein, u(1) represents the first target signal value collected at the first sampling moment, u(2) represents the first target signal value collected at the second moment, u(3) represents the first target signal value collected at the third moment, and so on, u(N) represents the first target signal value collected at the Nth moment;
[0162] The first target signal value has multiple possible results, each result is recorded as x, and the probability of each result is recorded as p(x), which is expressed as:
[0163]
[0164] The information entropy value corresponding to the first target signal is recorded as InEn(X) and is calculated as follows:
[0165] InEn(X)=-∑p(x)log2 p(x);
[0166] Here, num(u=x) represents the number of first target signal values x, and N represents the number of all first target signal values.
[0167] In the present application, when the early warning device obtains the four entropy values (sample entropy value, information entropy value, approximate entropy value, and fuzzy entropy value) mentioned above, it can also perform weighted calculations on the four entropy values to obtain corresponding weighted entropy values. That is to say, only the same entropy is used each time the entropy weighted calculation is performed.
[0168] Here, the four entropy values can be firstly calculated and processed for information entropy respectively to obtain the information entropy values corresponding to the four entropy values respectively, and then the collection end weight of the first target signal is determined based on the information entropy value, and the weighted entropy value of the entropy value is determined based on the collection end weight.
[0169] For example, M sensors are used to collect the first target signal, and the signal collection length is set to N. Among the first target signals collected continuously, every N first target signals are regarded as a group, recorded as Z, and expressed as:
[0170] Z=[Y(1),Y(2),Y(3),…,Y(K)];
[0171] Wherein, Y(1) represents the first group of first target signals collected, Y(2) represents the second group of first target signals collected, Y(3) represents the third group of first target signals collected, and so on, Y(K) represents the Kth group of first target signals collected;
[0172] Each group of first target signals can be recorded as Y, expressed as:
[0173] Y=[X(1),X(2),X(3),…,X(M)];
[0174] Wherein, X(1) represents the first target signal collected by the first sensor, X(2) represents the first target signal collected by the second sensor, X(3) represents the first target signal collected by the third sensor, and so on, X(M) represents the first target signal collected by the Mth sensor;
[0175] Each X can be expressed as:
[0176] X=[u(1),u(2),u(3),…,u(N)];
[0177] Wherein, u(1) represents the first target signal value collected at the first sampling moment, u(2) represents the first target signal value collected at the second sampling moment, u(3) represents the first target signal value collected at the third sampling moment, and so on, u(N) represents the first target signal value collected at the Nth sampling moment;
[0178] Then, for each group of first target data X under each sensor in Z, the sample entropy, information entropy, approximate entropy and / or fuzzy entropy values are calculated according to the above method, and the results can be uniformly expressed as:
[0179] En=[en(1),en(2),en(3),…,en(M);]
[0180] Wherein, en(1) represents all sample entropy values, information entropy values, approximate entropy values and / or fuzzy entropy values calculated based on the first target signal collected by the first sensor, en(2) represents all sample entropy values, information entropy values, approximate entropy values and / or fuzzy entropy values calculated based on the first target signal collected by the second sensor, and so on, en(M) represents all sample entropy values, information entropy values, approximate entropy values and / or fuzzy entropy values calculated based on the first target signal collected by the Mth sensor;
[0181] Here, I represents the first sensor; I is an integer between 1 and M; all sample entropy values, information entropy values, approximate entropy values and / or fuzzy entropy values calculated from the first target signal collected by the first sensor can be recorded as en(i), expressed as:
[0182] en(I)=[en(I,1),en(I,2),en(I,3),…,en(I,K)]
[0183] Wherein, en(I, 1) represents the sample entropy value, information entropy value, approximate entropy value and / or fuzzy entropy value calculated for the first group of first target signals collected under the I sensor, en(I, 2) represents the sample entropy value, information entropy value, approximate entropy value and / or fuzzy entropy value calculated for the second group of first target signals collected under the I sensor, and so on, en(I, K) represents the sample entropy value, information entropy value, approximate entropy value and / or fuzzy entropy value calculated for the Kth group of first target signals collected under the I sensor;
[0184] Next, calculate the weight corresponding to each acquisition end (such as microphone) of the first target signal:
[0185] First, the entropy values (including information entropy value, approximate entropy value, sample entropy value, and fuzzy entropy value) corresponding to the first target signal measured by different sensors in the same group are normalized, and the calculation formula is:
[0186]
[0187] Wherein, i represents the i-th sensor, i is an integer between 1 and M; j represents the j-th group of first target signals, j is an integer between 1 and K; gen(i, j) represents the entropy value corresponding to the j-th group of first target signals under the i-th sensor after normalization;
[0188] Then, calculate the proportion of the entropy value of each group measured by each sensor to the entropy value of all groups, denoted as p ij , the calculation formula is:
[0189]
[0190] Then calculate the information entropy of the entropy value of the group used under each sensor, expressed as E i , the calculation formula is:
[0191]
[0192] Among them, K represents the number of entropy values used to calculate the above information entropy under each sensor. If p ij =0, define E j =0. According to the above formula, the information entropy of the sample entropy of the working condition used by each sensor is calculated as E1, E2, ..., E M .
[0193] Then, based on the information entropy that has been calculated, the weight of each sensor is calculated:
[0194]
[0195] Among them, w i represents the weight of the i-th sensor, and M represents the number of sensors;
[0196] Here, the weighted entropy value corresponding to the next group of first target signals Y(K+1) of the Kth group of first target data Y(K) can be expressed as qEn(K+1), and the calculation formula is:
[0197] qEn(K+1)=En(1,K+1)*w1+En(2,K+1)*w2+…+En(M,K+1)*w M
[0198] Among them, qEn(K+1) is the weighted sample entropy value, weighted information entropy value, weighted approximate entropy value and / or weighted fuzzy entropy value corresponding to the next group of first target signals Y(K+1) of the Kth group of first target signals Y(K). That is to say, the data used to calculate the weighted entropy are all the entropy values or a fixed number of entropy values corresponding to the data group to be calculated. If the data group corresponds to sample entropy, all sample entropy values or a fixed number of sample entropy values corresponding to the data group are used to perform weighted entropy calculation; if the data group corresponds to information entropy, all information entropy values or a fixed number of information entropy values corresponding to the data group are used to perform weighted entropy calculation; if the data group corresponds to approximate entropy, all approximate entropy values or a fixed number of approximate entropy values corresponding to the data group are used to perform weighted entropy calculation; if the data group corresponds to fuzzy entropy, all fuzzy entropy values or a fixed number of fuzzy entropy values corresponding to the data group are used to perform weighted entropy calculation.
[0199] The present application further performs weighted calculation on the entropy values corresponding to the first target signal under different operating conditions, thereby improving the calculation accuracy of the result.
[0200] In the present application, the early warning device can also compare the weighted entropy value and / or entropy value of the compressor with a preset surge probability value; if the comparison result indicates that the difference between the weighted entropy value and / or entropy value and the surge probability value is less than a first threshold value, a first early warning signal is output, wherein the first early warning signal indicates that the surge state of the compressor is close to a surge boundary; if the comparison result indicates that the weighted entropy value and / or entropy value is greater than or equal to the surge probability value, a second early warning signal is output, wherein the second early warning signal indicates that the surge state of the compressor is that surge has occurred.
[0201] Here, different warning signals may also correspond to different surge probability values, and the larger the surge probability value, the higher the risk or possibility of compressor surge. For example, the surge probability value corresponding to the first warning signal is 80%, and the surge probability value corresponding to the second warning signal is 100%.
[0202] In the present application, the weighted entropy value and / or the entropy value may also change with changes in air parameters in the compressor.
[0203] The air parameters include but are not limited to the air flow rate of the compressed air inlet, the Mach number, the axial speed of the air inlet, and the flow coefficient.
[0204] The compressor surge warning method provided in the present application calculates the entropy value of the first target signal generated during the operation of the compressor and performs weighted calculation on the entropy value. The surge probability of the compressor is given an early warning by using the change law of the weighted entropy corresponding to the entropy value. This can greatly improve the warning accuracy and the safety performance of the compressor and avoid serious losses caused by surge of the compressor.
[0205] Figure 3 It is a curve diagram of the weighted entropy value of the aerodynamic noise time series signal at the intake end of the axial compressor, as shown in Figure 3 As shown:
[0206] Figure 3 The horizontal axis is the compressor mass flow rate, and the vertical axis is the weighted entropy value of the first target signal mentioned above. Figure 3 The middle curve describes the variation of the weighted entropy value of the first target signal with the compressor mass flow rate, that is, it presents an approximate W-shaped distribution. As the compressor mass flow rate decreases, the first target signal presents from right to left: the first decrease, the first increase, the second decrease, and the second increase.
[0207] Figure 4 The surge warning device of the compressor in this application is as follows: Figure 4 As shown, the device comprises:
[0208] A collection unit 401 is used to collect a first target signal generated by the compressor during operation, where the first target signal represents the operating state of the compressor;
[0209] A determination unit 402 is used to determine the entropy value corresponding to the first target signal under different operating conditions;
[0210] The processing unit 403 is used to perform weighted processing on the entropy value to obtain a weighted entropy value;
[0211] The output unit 404 is used to output a warning signal if the change state of the weighted entropy value and / or the entropy value meets a preset change rule, and the warning signal represents the surge state of the compressor.
[0212] In a preferred embodiment, the acquisition unit 401 is used to acquire the first target signal generated by the compressor during operation at intervals of preset time parameters based on a preset time window, wherein the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal; or, to continuously acquire the first target signal generated by the compressor during operation based on a preset time window, wherein the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal.
[0213] In a preferred embodiment, the determination unit 402 is used to determine the approximate entropy value corresponding to the first target signal under different operating conditions, and the first target signal is a time series signal under any operating condition.
[0214] In a preferred embodiment, the determination unit 402 is used to determine the fuzzy entropy value corresponding to the first target signal under different operating conditions, and the first target signal is a time series signal under any operating condition.
[0215] In a preferred embodiment, the determination unit 402 is used to determine the information entropy value corresponding to the first target signal under different operating conditions, and the first target signal is a time series signal under any operating condition.
[0216] In a preferred embodiment, the determination unit 402 is used to determine the sample entropy values corresponding to the first target signal under different operating conditions, and the first target signal is a time series signal under any operating condition.
[0217] In the preferred embodiment, the processing unit 403 is also used to process the entropy value to obtain the information entropy value of the entropy value; and determine the acquisition end weight of the first target signal based on the information entropy value; and determine the weighted entropy value of the entropy value based on the acquisition end weight.
[0218] Here, the weighted entropy value includes but is not limited to a weighted sample entropy value, a weighted approximate entropy value, a weighted information entropy value, and a weighted fuzzy entropy value.
[0219] In a preferred embodiment, the early warning device further comprises: a construction unit 405;
[0220] The construction unit 405 is used to construct a curve model based on the entropy values and / or weighted entropy values corresponding to the different operating conditions;
[0221] The determination unit 402 is also used to determine whether the change state of the entropy value and / or the weighted entropy value meets the preset change law if the curve model has at least one of the W-type and N-type structures, thereby triggering the output unit 504 to output a warning signal.
[0222] Here, the warning levels of the warning signals corresponding to the W type and the N type are different; wherein the warning level corresponding to the W type is higher than the warning level corresponding to the N type.
[0223] In the preferred embodiment, the early warning device further comprises: a comparison unit 406;
[0224] A comparison unit 406, configured to compare the weighted entropy value and / or the entropy value with a preset surge probability value;
[0225] If the comparison result indicates that the weighted entropy value and / or the difference between the entropy value and the surge probability value is less than a first threshold value, the output unit 404 outputs a first warning signal, wherein the first warning signal indicates that the surge state of the compressor is close to the surge boundary; if the comparison result indicates that the weighted entropy value and / or the entropy value is greater than or equal to the surge probability value, the output unit 404 outputs a second warning signal, wherein the second warning signal indicates that the surge state of the compressor is that surge has occurred.
[0226] Here, the weighted entropy value and / or the entropy value may vary as air parameters in the compressor vary.
[0227] It should be noted that the compressor surge warning device provided in the above embodiment is different from the above embodiment. Figure 1 and Figure 2 The provided compressor surge warning method belongs to the same concept, and the specific implementation process can refer to the above method embodiment, which will not be repeated here.
[0228] Figure 5 Schematic diagram of the surge warning device of the compressor in this application Figure 2 ,like Figure 5 As shown, the early warning device 500 includes at least one processor 501 and a memory 502 for storing a computer program that can be run on the processor 501, and the processor 501 is used to execute the surge early warning method suggested in the above embodiments of the present application when running the computer program. The early warning device 500 also includes at least one network interface 504 and a user interface 503. The various components in the early warning device 500 are coupled together through a bus system 505. It can be understood that the bus system 505 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 505 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, in Figure 5 Various buses are labeled as bus system 505 .
[0229] The user interface 503 may include a display, a keyboard, a mouse, a trackball, a click wheel, keys, buttons, a touch pad or a touch screen.
[0230] It can be understood that the memory 502 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAM bus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 502 described in the embodiments of the present application is intended to include but is not limited to these and any other suitable types of memories.
[0231] The memory 502 in the embodiment of the present application is used to store various types of data to support the operation of the early warning device 500. Examples of these data include: any computer program used to operate on the early warning device 500, such as an operating system 5021 and an application 5022; messages; pictures; videos, etc. Among them, the operating system 5021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application 5022 can include various applications, such as a media player (Media Player), a browser (Browser), etc., for implementing various application services. The program that implements the method of the embodiment of the present application can be included in the application 5022.
[0232] Processor 501 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in processor 501 or an instruction in software form. The above-mentioned processor 501 may be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Processor 501 can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application. A general-purpose processor may be a microprocessor or any conventional processor, etc. In conjunction with the steps of the method disclosed in the embodiments of the present application, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in a decoding processor. The software module may be located in a storage medium, which is located in memory 502, and processor 501 reads the information in memory 502, and completes the steps of the above method in conjunction with its hardware.
[0233] In an exemplary embodiment, the early warning device 500 can be implemented by one or more application-specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD), field-programmable gate array (FPGA), general-purpose processor, controller, microcontroller (MCU), microprocessor, or other electronic components to execute the aforementioned method.
[0234] In an exemplary embodiment, the present application also provides a computer-readable storage medium, such as a memory 502 including a computer program, and the computer program can be executed by the processor 501 of the early warning device 500 to complete the steps of the aforementioned method. The computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disk, or CD-ROM; it can also be various devices including one or any combination of the above memories, such as a computer, a tablet device, a personal digital assistant, etc.
[0235] A computer-readable storage medium stores a computer program, which, when executed by a processor, executes the compressor surge warning method suggested by the above embodiment of the present application.
[0236] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. In addition, the features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.
[0237] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A compressor surge warning method, characterized in that: The method comprises: Collecting a first target signal generated by the compressor during operation, wherein the first target signal represents an operating state of the compressor; Determining entropy values corresponding to the first target signal under different operating conditions; Performing weighted processing on the entropy value to obtain a weighted entropy value; If the weighted entropy value and / or the change state of the entropy value meets a preset change rule, a warning signal is output, and the warning signal represents the surge state of the compressor.
2. The method according to claim 1, characterized in that The collecting of the first target signal generated by the compressor during operation includes: collecting a first target signal generated by the compressor during operation at intervals according to preset time parameters based on a preset time window, wherein the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal; Or, a first target signal generated by the compressor during operation is continuously collected based on a preset time window, and the first target signal includes at least one of a dynamic pressure signal, a sound signal, and a vibration signal.
3. The method according to claim 1, characterized in that: The step of performing weighted processing on the entropy value to obtain a weighted entropy value includes: Processing the entropy value to obtain an information entropy value of the entropy value; Determine a collection end weight of the first target signal based on the information entropy value; A weighted entropy value of the entropy value is determined based on the acquisition end weight.
4. The method according to claim 1, characterized in that: The determining the entropy value corresponding to the first target signal under different operating conditions includes: Determine at least one of an information entropy value, a sample entropy value, a fuzzy entropy value, and an approximate entropy value corresponding to the first target signal under different operating conditions.
5. The method according to claim 4, characterized in that The first target signal is a time series signal under any operating condition, and determining the information entropy value corresponding to the first target signal under different operating conditions includes: The first target signal is recorded as X, which is expressed as: X=[u(1),u(2),u(3),…,u(N)] Wherein, u(1) represents the value of the first target signal collected at the first sampling moment, u(2) represents the value of the first target signal collected at the second moment, u(3) represents the value of the first target signal collected at the third moment, and so on, u(N) represents the value of the first target signal collected at the Nth moment; The value of the first target signal has multiple results, each result is recorded as x, and the probability of each result is recorded as p(x), which is expressed as: The information entropy value corresponding to the first target signal is denoted as InEn(X), and the calculation formula is: InEn(x)=-Σp(x)log2p(x); Here, num(u=x) represents the number of first target signal values x, and N represents the number of all first target signal values. The first target signal is a time series signal under any operating condition; determining the sample entropy value corresponding to the first target signal under different operating conditions includes: The first target signal is recorded as X, which is expressed as: X=[u(1),u(2),u(3),…,u(N)]; Wherein, u(1) represents the value of the first target signal collected at the first sampling moment, u(2) represents the value of the first target signal collected at the second moment, u(3) represents the value of the first target signal collected at the third moment, and so on, u(N) represents the value of the first target signal collected at the Nth moment; Any subsequence of the first target signal starting from the i-th sampling time is denoted as X(i) and is expressed as: X(i)=[yu(i),u(i+1),…,u(i+m-1)]; Wherein, i represents the i-th sampling moment, i is any integer between 1 and N-m+1; u(i) is the value of the first target signal collected at the i-th sampling moment; m is the preset sequence dimension value, ranging from 1 to Ni; Under the condition of determining the value of i, traverse all combinations of X(i) and X(j) and calculate the maximum distance between X(i) and X(j), which is recorded as: d[X(i), X(j)], and the calculation formula is: d[X(i),X(j)]=max k=0....,m-1 |u(i+k)-u(j+k)|; Wherein, X(i) represents any subsequence of the first target signal starting from the i-th sampling moment; X(j) represents any subsequence of the first target signal starting from the j-th sampling moment, i is not equal to j; u(i+k) and u(j+k) represent the values of the first target signal collected at the i+k-th sampling moment and the j+k-th sampling moment, respectively; k is any integer between 0 and m-1; max represents the maximum value of all |u(i+k)-u(j+k)| calculated from k from 0 to m-1; Based on the distance between X(i) and X(j), we count the number of items between X(i) and X(j) that meet the requirements in 1≤i≤Nm and 1≤j≤Nm and j≠i, which is denoted as B i (r), the calculation formula is: B i (r)=num{d[X(i),X(j)]≤r}; The condition requires that the distance between X(i) and X(j) (d[X(i), X(j)]) is less than or equal to the preset parameter r value, and each d[X(i), X(j)] is regarded as 1 item, and the items that meet the condition are recorded as 1, and the items that do not meet the condition are recorded as 0; Find the ratio of the number of items that meet the conditional requirements to the total number, denoted as The calculation formula is: Then find the logarithmic mean of the ratio, denoted as Φ m (r), the calculation formula is: Among them, Nm means that under the condition of determining the value of i, all combinations of X(i) and X(j) are traversed, and there are Nm combinations, that is, the total number; The sample entropy value corresponding to the first target signal is denoted as SaEn, and the calculation formula is: Among them, the sequence dimension m is added by 1 to get m+1, and Φ m (r) and Φ m+1 (r) is obtained in the same way.
6. The method according to claim 1, characterized in that If the weighted entropy value and / or the change state of the entropy value meets the preset change rule, outputting a warning signal includes: Constructing a curve model based on the weighted entropy value and / or the entropy value corresponding to the different operating conditions; If the curve model presents at least one of a W-type and an N-type structure, it is determined that the weighted entropy value and / or the change state of the entropy value satisfies a preset change rule, and a warning signal is output.
7. The method according to claim 6, characterized in that The warning levels of the warning signals corresponding to the W type and the N type are different; wherein the warning level corresponding to the W type is higher than the warning level corresponding to the N type.
8. The method according to claim 6, characterized in that The method further comprises: comparing the weighted entropy value and / or the entropy value with a preset surge probability value; If the comparison result indicates that the weighted entropy value and / or the difference between the entropy value and the surge probability value is less than a first threshold, outputting a first warning signal, wherein the first warning signal indicates that the surge state of the compressor is close to a surge boundary; If the comparison result indicates that the weighted entropy value and / or the entropy value is greater than or equal to the surge probability value, a second warning signal is output, wherein the second warning signal indicates that the surge state of the compressor is that surge has occurred.
9. A compressor surge warning device, characterized in that: The device comprises: A collection unit, used for collecting a first target signal generated by the compressor during operation, wherein the first target signal represents an operating state of the compressor; A determination unit, configured to determine an entropy value corresponding to the first target signal under different operating conditions; A processing unit, used for performing weighted processing on the entropy value to obtain a weighted entropy value; The output unit is used to output a warning signal if the weighted entropy value and / or the change state of the entropy value meets a preset change rule, and the warning signal represents the surge state of the compressor.
10. A compressor surge warning device, characterized in that: The device comprises: a processor and a memory for storing a computer program that can be run on the processor, and the processor is configured to execute the surge warning method according to any one of claims 1 to 8 when running the computer program.
