Cable laying tension and pressure signal time-frequency dynamic analysis system

Abstract

The invention relates to the technical field of signal analysis, in particular to a cable laying tension and pressure signal time-frequency dynamic analysis system, which comprises a signal acquisition module used for acquiring tension and pressure signals in cable laying; the signal analysis module is used for acquiring time domain features and frequency domain features of the tension and pressure signals and constructing a data prediction model; the construction quality evaluation module is used for evaluating the construction quality according to the construction experience, the time domain characteristics of the tension and pressure signals and the frequency domain characteristics of the tension and pressure; and the monitoring alarm module is used for judging whether threshold values are exceeded or not according to the output content of the data prediction model and the time domain characteristics and the frequency domain characteristics of the tension and pressure signals. According to the invention, by analyzing the time domain and the frequency domain of various tension and pressure signals and using the data prediction unit to construct the data prediction model, various states of the power cable in the construction process can be known in time, the cable laying quality is improved, and the service life of the cable is prolonged.

Description

Technical Field

[0001] The invention relates to the technical field of signal analysis, in particular to a time-frequency dynamic analysis system for cable laying tension and pressure signals. Background Art

[0002] The development of power cable construction technology is a process of continuous enhancement of mechanization and intelligence. In the early days, cable laying construction relied entirely on manual traction and laying. As the cross-sectional area of ​​the cable increased, the weight also increased, and greater traction was required for laying. Therefore, mechanical equipment that can provide great traction has been widely used. The laying of high-voltage or larger cross-section cables requires more advanced mechanical laying methods. At the same time, how to avoid mechanical damage to the cable during the laying process and how to effectively monitor the laying process are new issues facing cable laying.

[0003] Mechanized construction of power cables can realize the transformation of cable line construction from labor-intensive to equipment-intensive, thereby improving construction efficiency and meeting the needs of large-scale power grid construction; improving the process quality level of power cable construction; At present, mechanized construction of power cables has reached a certain scale and level, but intelligent construction is still in its infancy. In the intelligent construction of power cables, how to combine online monitoring equipment to judge the various states of power cables during the construction process is particularly important. Summary of the invention

[0004] The purpose of the present invention is to provide a time-frequency dynamic analysis system for cable laying tension and pressure signals to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a time-frequency dynamic analysis system for cable laying tension and pressure signals, comprising:

[0006] Signal acquisition module: used to collect the tension pressure signal during cable laying and transmit it with low noise;

[0007] Signal analysis module: used to obtain the time domain characteristics and frequency domain characteristics of the tension and pressure signals, and to build a data prediction model. The output of the data prediction model is the potential problems and dynamic characteristics in the laying process. The dynamic characteristics include time domain characteristics and frequency domain characteristics. The potential problem types are inferred based on the dynamic characteristics.

[0008] Construction quality assessment module: used to assess construction quality based on construction experience, time domain characteristics of tension and pressure signals, and frequency domain characteristics of tension and pressure;

[0009] Monitoring and alarm module: used to determine whether the threshold is exceeded based on the output content of the data prediction model, the time domain characteristics and frequency domain characteristics of the tension and pressure signals.

[0010] As a further improvement of the present technical solution, the signal acquisition module includes a signal acquisition unit, a low-noise amplification unit, a low-noise power supply unit, an analog-to-digital conversion unit, a low-pass power frequency suppression unit, a sampling unit and a transmission unit, and the signal acquisition unit, the low-noise amplification unit, the low-pass power frequency suppression unit, the sampling unit and the transmission unit are electrically connected in sequence.

[0011] As a further improvement of the present technical solution, the signal acquisition unit includes a plurality of pressure sensors, and the signals collected by the signal acquisition unit include the driving force of the cable unwinding machine, the cable side pressure of the cable unwinding machine, the conveying speed, the cable contact force and the cable traction force. The low-noise amplification unit, the low-pass power frequency suppression unit, the sampling unit and the transmission unit are all composed of FPGA. The analog-to-digital conversion unit realizes the analog-to-digital conversion function of multiple 24-bit resolution differential ADCs in parallel at a rate of 10kHz through FPGA. The low-pass power frequency suppression unit realizes the function of low-pass power frequency suppression by forming a digital low-pass filter and a digital notch filter through FPGA. The sampling unit adopts high-resolution oversampling technology, and the transmission unit is signal-connected to the signal analysis module.

[0012] As a further improvement of the present technical solution, the signal analysis module includes a time domain analysis unit, a frequency domain analysis unit and a data prediction unit. The time domain analysis unit obtains the time domain characteristics of the signal, specifically including maximum amplitude, average value, variance, power and energy integral parameters. The frequency domain analysis unit obtains the frequency domain characteristics of the signal based on Fourier transform.

[0013] As a further improvement of the technical solution, the time domain analysis unit obtains the characteristics of the signal x(n) based on the following formula:

[0014] A amx =max(∣x(n)∣),n=0,1,…,N-1

[0015]

[0016]

[0017] Where n is the time index, N is the number of sampling points, max is the maximum value operation, |x(n)| is the absolute value operation of x(n), and A amx is the maximum value, is the mean value, σ 2 is the variance, P is the power, and E is the energy;

[0018] The time domain analysis unit is also used to analyze the time domain correlation. For two signals x(n) and y(n), there is:

[0019]

[0020] Where n is the time index, R x，y (m) is the correlation function value between signals x(n) and y(n), m is the time delay, n is the time index, indicating the sampling point position of the signal, and y(nm) is the signal y(n) after m units of time delay.

[0021] As a further improvement of the technical solution, the formula for the frequency domain analysis unit to obtain the spectrum characteristics of the signal x(n) is:

[0022]

[0023] E(k)=|X(k)| 2

[0024] Where n is the time index, k is the frequency index, N is the number of sampling points, j is the imaginary unit, X(k) is the spectral component of the signal x(n) at k, E(k) is the energy of the signal x(n) at k, and |X(k)| is the absolute value of X(k).

[0025] As a further improvement of the present technical solution, the data prediction unit constructs a data prediction model based on the optimized random forest, uses a historical annotation data set as a data set, and uses the time domain features and frequency domain features of the historical annotation data set obtained by the time domain analysis unit and the frequency domain analysis unit as input data. The output content of the data prediction model is potential problems and dynamic characteristics in the laying process, and the improved bat algorithm is used to optimize the parameters in the random forest. The specific parameters include: the number of decision trees, the maximum depth of the decision tree, the minimum number of samples required to split internal nodes, the minimum number of samples selected by leaf nodes, and the maximum number of features required to construct a decision tree.

[0026] As a further improvement of the technical solution, the specific steps of using the improved bat algorithm to optimize the parameters in the random forest in the data prediction unit are:

[0027] Step 1: Population initialization, including the number of individuals in the population, the initial position of the individuals, the spatial dimension, the maximum number of iterations, the initial speed, the upper and lower limits of the pulse frequency, the pulse loudness and the pulse emission rate. The spatial dimension is five, and the position of the individuals in different dimensions represents the parameters to be optimized.

[0028] Step 2: Calculate the fitness function and use the accuracy of the random forest model as the objective function;

[0029] Step 3: Update position and speed. The corresponding formula is:

[0030] f i =f min +β(f ma -fmin )

[0031] v i (t+1)=w i (t)v i (t)+(x i (t)-x best (t))f i

[0032] x i (t+1)=x i (t)+v i (t+1)

[0033]

[0034] Where t is the number of iterations, f min is the lower limit of pulse frequency, f max is the upper limit of the pulse frequency, β is a random number uniformly distributed within (0, 1), f i is the pulse frequency of the ith bat, v i (t+1) is the speed of the ith bat at the (t+1)th generation, v i (t) is the speed of the ith bat in the tth generation, x i (t+1) is the position of the ith bat in the (t+1)th generation, x i (t) is the position of the ith bat in the tth generation, x best (t) is the optimal bat position in the tth generation, w i (t) is the weight of the ith bat in the tth generation, w max Set the maximum value of the weight, w min is the minimum value set for the weight, t max is the maximum number of iterations;

[0035] Step 4: Generate random numbers rand1 and rand2. Both rand1 and rand2 are random numbers on [0, 1]. If rand1>r i (t), then choose to locally perturb the current optimal solution and update it according to the following formula:

[0036]

[0037] Among them, r i (t) is the pulse emission rate of the ith bat in the tth generation, x new (t) is the newly generated solution, x best (t) is the position of the optimal bat in the tth generation, ε is a random number on [-1,1], is the average loudness of the tth generation of the group;

[0038] The new solution is accepted when the following equation is satisfied:

[0039] f new (t)>f best (t)and rand2 i (t)

[0040] Among them, f new (t) is the new solution x new (t) The corresponding fitness value, f best (t) is the optimal bat x best (t) The corresponding fitness value, A i (t) is the pulse loudness of the tth generation of the i-th bat;

[0041] A i (t+1) = α A i (t)

[0042] r i (t+1)=r i (0)(1-exp(-γt))

[0043] Among them, A i (t+1) is the impulse loudness of the ith bat at the (t+1)th generation, α and γ are both preset parameters, and 0<α<1, γ>0, r i (0) is the initial pulse emission rate of the i-th bat, r i (t+1) is the pulse emission rate of the i-th bat at the (t+1)th generation;

[0044] Step 5: Introduce Cauchy mutation. In each iteration, perform Cauchy mutation on some individuals, and we have:

[0045]

[0046] in, is the individual position after mutation, x is the individual position to be mutated, Cauch y(0,1) is a random number that conforms to the standard Cauchy distribution;

[0047] Step 6: Introduce the elimination mechanism. In each iteration, eliminate the P individuals with the lowest fitness and generate P new individuals:

[0048]

[0049] in, is the position of the newly generated individual, x best is the position of the best bat in the current population, tanh is the bitangent function, and rand3 is a random number in (0, 1);

[0050] ​Step 7: Determine the termination condition. The process terminates when the number of iterations reaches or the set number of iterations output the same result.

[0051] As a further improvement of the present technical solution, the quantitative performance of the construction quality assessment module includes maximum tensile and compressive force, allowable range of high-frequency tensile and compressive force distribution, tensile and compressive force energy distribution range and the area of ​​maximum correlation between the two forces, and the construction quality is assessed based on the quantitative performance.

[0052] As a further improvement of the present technical solution, the monitoring and alarm module sets thresholds for the driving force of the cable unwinding machine, the cable side pressure of the cable unwinding machine, the conveying speed, the cable contact force and the cable traction force, and also sets thresholds for the output content of the data prediction model, and alarms when the thresholds are exceeded.

[0053] Compared with the prior art, the beneficial effects of the present invention are as follows: in the cable laying tension and pressure signal time-frequency dynamic analysis system, by analyzing the time domain and frequency domain of multiple tension and pressure signals and using a data prediction unit to build a data prediction model, it is helpful to timely understand the various states of the power cable during the construction process, and can effectively alarm for abnormal situations, thereby improving the cable laying quality and cable service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is the overall structure diagram of the present invention;

[0055] Figure 2 It is the overall structure diagram of the signal acquisition module in the present invention;

[0056] Figure 3 It is the overall structure diagram of the signal analysis module in the present invention.

[0057] Figure numerals: 10, signal acquisition module; 100, signal acquisition unit; 101, low-noise amplification unit; 102, low-noise power supply unit; 103, analog-to-digital conversion unit; 104, low-pass power frequency suppression unit; 105, sampling unit; 106, transmission unit; 20, signal analysis module; 200, time domain analysis unit; 201, frequency domain analysis unit; 202, data prediction unit; 30, construction quality assessment module; 40, monitoring and alarm module. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] This embodiment provides a time-frequency dynamic analysis system for cable laying tension and pressure signals, such as Figure 1-Figure 3 As shown, including:

[0060] Signal acquisition module 10: used to collect the tension pressure signal during cable laying and transmit it with low noise. It uses a highly sensitive sensor and the low noise transmission can effectively eliminate external interference.

[0061] Signal analysis module 20: used to obtain the time domain characteristics and frequency domain characteristics of the tension and pressure signals, and to build a data prediction model. The output of the data prediction model is the potential problems and dynamic characteristics in the laying process. The dynamic characteristics include time domain characteristics and frequency domain characteristics. The potential problem types are inferred based on the dynamic characteristics. A preset rule base is used to infer the potential problem types based on the dynamic characteristics and the rule base. This allows for a more detailed analysis of the quality of the cable laying process, providing guidance for smooth construction and quality control.

[0062] Construction quality assessment module 30: used to assess the construction quality according to the time domain characteristics of the tension and pressure signals and the frequency domain characteristics of the tension and pressure signals, fully combining the on-site construction experience, and carefully quantifying the assessment of the construction quality;

[0063] Monitoring and alarm module 40: used to determine whether the threshold is exceeded based on the output content of the data prediction model, the time domain characteristics and frequency domain characteristics of the tension and pressure signals, and ensure the normal progress of the construction through multiple thresholds.

[0064] The signal acquisition module 10 includes a signal acquisition unit 100, a low-noise amplifier unit 101, a low-noise power supply unit 102, an analog-to-digital conversion unit 103, a low-pass power frequency suppression unit 104, a sampling unit 105 and a transmission unit 106. The signal acquisition unit 100, the low-noise amplifier unit 101, the low-pass power frequency suppression unit 104, the sampling unit 105 and the transmission unit 106 are electrically connected in sequence, and the low-noise power supply unit 102 provides power for other units.

[0065] The signal acquisition unit 100 includes several pressure sensors. The signals collected by the signal acquisition unit 100 include the driving force of the cable unwinding machine, the pressure on the cable side of the cable unwinding machine, the conveying speed, the cable contact force and the cable traction force. The low-noise amplification unit 101, the low-pass power frequency suppression unit 104, the sampling unit 105 and the transmission unit 106 are all composed of FPGA. The analog-to-digital conversion unit 103 realizes the analog-to-digital conversion function of multiple 24-bit resolution differential ADCs at a parallel rate of 10kHz through FPGA. The low-pass power frequency suppression unit 104 realizes the function of low-pass power frequency suppression by forming a digital low-pass filter and a digital notch filter through FPGA. The sampling unit 105 adopts high-resolution oversampling technology. The transmission unit 106 is connected to the signal analysis module 20 by signal. The signal acquisition unit 100 can suppress environmental high-frequency noise and mains interference involving various AC power in the workplace. The low-pass power frequency suppression unit 104 and the sampling unit 105 cooperate with filtering means to suppress out-of-band interference and in-band specific frequency interference, thereby further improving the signal-to-noise ratio.

[0066] The signal analysis module 20 includes a time domain analysis unit 200, a frequency domain analysis unit 201 and a data prediction unit 202. The time domain analysis unit 200 obtains the time domain characteristics of the signal, specifically including the maximum amplitude, mean value, variance, power and energy integration parameters. The frequency domain analysis unit 201 obtains the frequency domain characteristics of the signal based on Fourier transform, and deeply displays the intrinsic characteristics of the signal from the time domain and frequency domain dimensions, thereby revealing the physical characteristics behind the signal, and can achieve a more detailed analysis of the quality of the cable laying process.

[0067] The time domain analysis unit 200 obtains the characteristics of the signal x(n) based on the following formula:

[0068] A amx =max(∣x(n)∣),n=0,1,…,N-1

[0069]

[0070]

[0071] Where n is the time index, N is the number of sampling points, max is the maximum value operation, |x(n)| is the absolute value operation of x(n), and A amx is the maximum value, is the mean value, σ 2 is the variance, P is the power, and E is the energy.

[0072] The time domain analysis unit 200 is also used to analyze the time domain correlation. For two signals x(n) and y(n), there is:

[0073]

[0074] Among them, Rx，y (m) is the correlation function value between signals x(n) and y(n), m is the time delay, n is the time index, indicating the sampling point position of the signal, y(nm) is the signal y(n) after m units of time delay. For example, the correlation function calculation between the cable traction force and the cable side pressure signal of the cable arrangement machine and the correlation function calculation results of different time spans can be given.

[0075] The formula for the frequency domain analysis unit 201 to obtain the frequency spectrum characteristics of the signal x(n) is:

[0076]

[0077] E(k)=|X(k)| 2

[0078] Among them, n is the time index, k is the frequency index, N is the number of sampling points, j is the imaginary unit, X(k) is the spectral component of the signal x(n) at k, E(k) is the energy of the signal x(n) at k, and |X(k)| is the absolute value of X(k). For example, the frequency domain decomposition results of the cable traction force and the cable side pressure of the cable arranger can be given, and the energy distribution of the cable traction force and the cable side pressure of the cable arranger can be displayed at the same time.

[0079] The data prediction unit 202 constructs a data prediction model based on the optimized random forest, uses the historical annotation data set as the data set, and uses the time domain features and frequency domain features obtained by the time domain analysis unit 200 and the frequency domain analysis unit 201 of the historical annotation data set as input data. The output content of the data prediction model is the potential problems and dynamic characteristics in the laying process, and uses the improved bat algorithm to optimize the parameters in the random forest. The specific parameters include: the number of decision trees, the maximum depth of the decision tree, the minimum number of samples required to split the internal nodes, the minimum number of samples selected by the leaf nodes, and the maximum number of features required to build a decision tree. 70% of the data set is used as a training set, and the rest is used as a validation set to build a random forest. The Bootstrap algorithm is used to extract a number of samples from the training set with replacement to generate a number of decision trees. When splitting each node of the decision tree, a number of attributes are randomly extracted from all the attributes, and a decision tree is constructed according to the Gini index and the classification and regression tree algorithm.

[0080] The specific steps of using the improved bat algorithm to optimize the parameters in the random forest in the data prediction unit 202 are:

[0081] Step 1: Population initialization, including the number of individuals in the population, the initial position of the individuals, the spatial dimension, the maximum number of iterations, the initial speed, the upper and lower limits of the pulse frequency, the pulse loudness, and the pulse emission rate. The spatial dimension is five, and the positions of the individuals in different dimensions represent the parameters to be optimized, namely, the number of decision trees, the maximum depth of the decision tree, the minimum number of samples required to split the internal nodes, the minimum number of samples selected by the leaf nodes, and the maximum number of features required to construct the decision tree;

[0082] Step 2: Calculate the fitness function and use the accuracy of the random forest model as the objective function;

[0083] Step 3: Update position and speed. The corresponding formula is:

[0084] f i =f min +β(f max -f min )

[0085] v i (t+1)=w i (t)v i (t)+(x i (t)-x best (t))f i

[0086] x i (t+1)=x i (t)+v i (t+1)

[0087]

[0088] Where t is the number of iterations, f min is the lower limit of pulse frequency, f max is the upper limit of the pulse frequency, β is a random number uniformly distributed within (0, 1), f i is the pulse frequency of the ith bat, v i (t+1) is the speed of the ith bat at the (t+1)th generation, v i (t) is the speed of the ith bat in the tth generation, x i (t+1) is the position of the ith bat in the (t+1)th generation, x i (t) is the position of the ith bat in the tth generation, x best (t) is the optimal bat position in the tth generation, w i (t) is the weight of the ith bat in the tth generation, w max Set the maximum value of the weight, w min is the minimum value set for the weight, t maxFor the maximum number of iterations, an adaptive weight is introduced to balance the global search capability and the local search capability;

[0089] Step 4: Generate random numbers rand1 and rand2. Both rand1 and rand2 are random numbers on [0, 1]. If rand1>r i (t), then choose to locally perturb the current optimal solution and update it according to the following formula:

[0090]

[0091] Among them, r i (t) is the pulse emission rate of the ith bat in the tth generation, x new (t) is the newly generated solution, x best (t) is the position of the optimal bat in the tth generation, ε is a random number on [-1,1], is the average loudness of the tth generation of the group;

[0092] The new solution is accepted when the following equation is satisfied:

[0093] f new (t)>f best (t)and rand2 i (t)

[0094] Among them, f new (t) is the new solution x new (t) The corresponding fitness value, f best (t) is the optimal bat x best (t) The corresponding fitness value, A i (t) is the pulse loudness of the tth generation of the i-th bat;

[0095] A i (t+1) = α A i (t)

[0096] r i (t+1)=r i (0)(1-exp(-γt))

[0097] Among them, A i (t+1) is the impulse loudness of the ith bat at the (t+1)th generation, α and γ are both preset parameters, and 0<α<1, γ>0, r i (0) is the initial pulse emission rate of the i-th bat, r i (t+1) is the pulse emission rate of the i-th bat at the (t+1)th generation;

[0098] ​Step 5: Introducing Cauchy mutation can increase diversity in the search process, avoid premature convergence to the local optimal solution, and increase the possibility of finding the global optimal solution. In each iteration, Cauchy mutation operation is performed on some individuals, and there are:

[0099]

[0100] in, is the individual position after mutation, x is the individual position to be mutated, Cauchh y(0,1) is a random number that conforms to the standard Cauchy distribution;

[0101] Step 6: Introduce an elimination mechanism to improve the convergence speed. In each iteration, eliminate the P individuals with the lowest fitness and generate P new individuals:

[0102]

[0103] in, is the position of the newly generated individual, x best is the position of the best bat in the current population, tanh is the bitangent function, and rand3 is a random number in (0, 1);

[0104] Step 7: Determine the termination condition. The process terminates when the maximum number of iterations is reached or the set number of iterations output the same result.

[0105] The quantitative performance of the construction quality assessment module 30 includes the maximum tension and pressure, the allowable range of high-frequency tension and pressure distribution, the tension and pressure energy distribution range and the maximum correlation area between the two forces. The construction quality is evaluated based on the quantitative performance, and the signal characteristic law under high-quality construction that has been verified and recognized by the industry is given, thereby defining the time-frequency quantitative definition of the cable laying construction quality. The monitoring and alarm module 40 sets the thresholds of the cable laying machine driving force, the cable side pressure of the cable laying machine, the conveying speed, the cable contact force and the cable traction force. It also sets the threshold of the output content of the data prediction model, and alarms when the threshold is exceeded, so that abnormal situations that occur during construction can be discovered in time and responded to in time.

[0106] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. Cable laying tension and pressure signal time-frequency dynamic analysis system, characterized by: include: Signal acquisition module: used to collect the tension pressure signal during cable laying and transmit it with low noise; Signal analysis module: used to obtain the time domain characteristics and frequency domain characteristics of the tension and pressure signals, and to build a data prediction model. The output of the data prediction model is the potential problems and dynamic characteristics in the laying process. The dynamic characteristics include time domain characteristics and frequency domain characteristics. The potential problem types are inferred based on the dynamic characteristics. Construction quality assessment module: used to assess construction quality based on the time domain characteristics of tension and pressure signals and the frequency domain characteristics of tension and pressure; Monitoring and alarm module: used to determine whether the threshold is exceeded based on the output content of the data prediction model, the time domain characteristics and frequency domain characteristics of the tension and pressure signals.

2. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 1 is characterized by: The signal acquisition module includes a signal acquisition unit, a low-noise amplification unit, a low-noise power supply unit, an analog-to-digital conversion unit, a low-pass power frequency suppression unit, a sampling unit and a transmission unit. The signal acquisition unit, the low-noise amplification unit, the low-pass power frequency suppression unit, the sampling unit and the transmission unit are electrically connected in sequence.

3. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 2 is characterized by: The signal acquisition unit includes several pressure sensors. The signals collected by the signal acquisition unit include the driving force of the cable unwinding machine, the cable side pressure of the cable unwinding machine, the conveying speed, the cable contact force and the cable traction force. The low-noise amplification unit, the low-pass power frequency suppression unit, the sampling unit and the transmission unit are all composed of FPGA. The analog-to-digital conversion unit realizes the analog-to-digital conversion function of multiple 24-bit resolution differential ADCs at a parallel rate of 10kHz through FPGA. The low-pass power frequency suppression unit realizes the function of low-pass power frequency suppression by forming a digital low-pass filter and a digital notch filter through FPGA. The sampling unit adopts high-resolution oversampling technology, and the transmission unit is signal-connected to the signal analysis module.

4. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 1 is characterized by: The signal analysis module includes a time domain analysis unit, a frequency domain analysis unit and a data prediction unit. The time domain analysis unit obtains the time domain characteristics of the signal, specifically including maximum amplitude, average value, variance, power and energy integration parameters. The frequency domain analysis unit obtains the frequency domain characteristics of the signal based on Fourier transform.

5. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 4 is characterized in that: The time domain analysis unit obtains the characteristics of the signal x(n) based on the following formula: A amx =max(∣x(n)∣),n=0,1,…,N-1 Where n is the time index, N is the number of sampling points, max is the maximum value operation, |x(n)| is the absolute value operation of x(n), and A amx is the maximum value, is the mean value, σ 2 is the variance, p is the power, and E is the energy; The time domain analysis unit is also used to analyze the time domain correlation. For two signals x(n) and y(n), there is: Where n is the time index, R x，y (m) is the correlation function value between signals x(n) and y(n), m is the time delay, n is the time index, indicating the sampling point position of the signal, and y(nm) is the signal y(n) after m units of time delay.

6. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 4 is characterized in that: The frequency domain analysis unit obtains the frequency spectrum characteristics of the signal x(n) as follows: E(k)=∣X(k)∣ 2 Where n is the time index, k is the frequency index, N is the number of sampling points, j is the imaginary unit, X(k) is the spectral component of the signal x(n) at k, E(k) is the energy of the signal x(n) at k, and |X(k)| is the absolute value of X(k).

7. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 4 is characterized by: The data prediction unit constructs a data prediction model based on the optimized random forest, uses a historical annotation data set as a data set, and uses the time domain features and frequency domain features of the historical annotation data set obtained by the time domain analysis unit and the frequency domain analysis unit as input data. The output content of the data prediction model is potential problems and dynamic characteristics in the laying process, and uses the improved bat algorithm to optimize the parameters in the random forest. The specific parameters include: the number of decision trees, the maximum depth of the decision tree, the minimum number of samples required to split the internal nodes, the minimum number of samples selected by the leaf node, and the maximum number of features required to construct the decision tree.

8. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 7 is characterized in that: The specific steps of using the improved bat algorithm to optimize the parameters in the random forest in the data prediction unit are: Step 1: Population initialization, including the number of individuals in the population, the initial position of the individuals, the spatial dimension, the maximum number of iterations, the initial speed, the upper and lower limits of the pulse frequency, the pulse loudness and the pulse emission rate. The spatial dimension is five, and the position of the individuals in different dimensions represents the parameters to be optimized. Step 2: Calculate the fitness function and use the accuracy of the random forest model as the objective function; Step 3: Update position and speed. The corresponding formula is: f i =f min +β(f max -f min ) v i (t+1)=w i (t)v i (t)+(x i (t)-x best (t))f i x i (t+1)=x i (t)+v i (t+1) Where t is the number of iterations, f min is the lower limit of pulse frequency, f max is the upper limit of the pulse frequency, β is a random number uniformly distributed within (0,1), f i is the pulse frequency of the ith bat, v i (t+1) is the speed of the ith bat at the (t+1)th generation, v i (t) is the speed of the ith bat in the tth generation, x i (t+1) is the position of the ith bat in the (t+1)th generation, x i (t) is the position of the ith bat in the tth generation, x best (t) is the optimal bat position in the tth generation, w i (t) is the weight of the ith bat in the tth generation, w max Set the maximum value of the weight, w men is the minimum value set for the weight, t max is the maximum number of iterations; Step 4: Generate random numbers rand1 and rand2. Both rand1 and rand2 are random numbers on [0, 1]. If rand1>r i (t), then choose to locally perturb the current optimal solution and update it according to the following formula: Among them, r i (t) is the pulse emission rate of the ith bat in the tth generation, x new (t) is the newly generated solution, x best (t) is the position of the optimal bat in the tth generation, ε is a random number on [-1, 1], is the average loudness of the tth generation of the group; The new solution is accepted when the following equation is satisfied: f new (t)>f best (t) and rand2<A i (t) Among them, f new (t) is the new solution x new (t) The corresponding fitness value, f best (t) is the optimal bat x best (t) The corresponding fitness value, A i (t) is the pulse loudness of the tth generation of the i-th bat; A i (t+1)=αA i (t) r i (t+1)=r i (0)(1-exp(-γt)) Among them, A i (t+1) is the impulse loudness of the ith bat at the (t+1)th generation, α and γ are both preset parameters, and 0<α<1, γ>0, r i (0) is the initial pulse emission rate of the i-th bat, r i (t+1) is the pulse emission rate of the i-th bat at the (t+1)th generation; Step 5: Introduce Cauchy mutation. In each iteration, perform Cauchy mutation on some individuals, and we have: in, is the individual position after mutation, x is the individual position to be mutated, Cauchy(0,1) is a random number that conforms to the standard Cauchy distribution; Step 6: Introduce the elimination mechanism. In each iteration, eliminate the P individuals with the lowest fitness and generate P new individuals: in, is the position of the newly generated individual, x best is the position of the best bat in the current population, tanh is the double tangent function, and rand3 is a random number in (0,1); Step 7: Determine the termination condition. The process terminates when the number of iterations reaches or the set number of iterations output the same result.

9. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 1 is characterized by: The quantitative performance of the construction quality assessment module includes maximum tension and pressure, allowable range of high-frequency tension and pressure distribution, tension and pressure energy distribution range and maximum correlation area between the two forces, and the construction quality is assessed based on the quantitative performance.

10. The cable laying tension and pressure signal time-frequency dynamic analysis system according to claim 1, characterized in that: The monitoring alarm module sets thresholds for the driving force of the cable unwinding machine, the cable side pressure of the cable unwinding machine, the conveying speed, the cable contact force and the cable traction force, and also sets thresholds for the output content of the data prediction model, and alarms when the thresholds are exceeded.

Images