The concept of load sharing: In multi-axis mechanical systems, especially when the axes are connected by the materials being processed, the motors on different axes may be in different operating states. One motor may be in a motoring state, while another may be in a generator state. The motor operating in a motoring state draws electrical energy from its power supply, while the motor operating in a generator state outputs electrical energy back to its power supply; this energy is also called "regenerative energy." We can see from the motor's torque characteristic diagram that these two operating modes operate in different regions. As shown in the diagram, when Nx > N0, the motor's output torque is opposite to the direction of motion, and the output torque becomes a braking force, preventing the motor speed from increasing further. The crane's main hook operates in this state when descending with a load. At this time, the motor speed is already greater than the synchronous speed; in fact, the motor is being dragged along, and the motor generates "regenerative energy" to feed back energy to the grid. Because the frequency and phase of the feedback potential are not exactly the same as those of the grid, this is not allowed in most cases. Currently, China only allows oilfield systems to feed back electrical energy to the grid. Traditional PWM inverters are not designed to feed regenerative energy back to the three-phase power supply. Therefore, all energy absorbed from the motor is fed back into the electrolytic capacitors, ultimately increasing the DC bus voltage in the inverter. If the inverter is equipped with a braking unit and braking resistor, the resistor can be connected, allowing the regenerative energy to be dissipated as heat. Of course, the power of the braking unit and braking resistor must be considered, and the energy must be dissipated using high-frequency pulses to maintain bus voltage balance. Let's examine how the regenerative energy is fed back to the DC bus. Most modern inverters use IGBT (Insulated Gate Bipolar Transistor) modules for their inverter units. To protect the IGBTs, reverse diodes are connected to both the collector and emitter terminals, as shown in the diagram. When the motor is generating power, the instantaneous terminal voltage of the motor is greater than the DC bus voltage. At this point, the IGBT is under reverse voltage, so the reverse diodes conduct, forming a circuit from the two reverse diodes in different phases, charging the capacitors on the DC bus and ultimately increasing the DC bus voltage. In multi-motor linkage, one of the frequency converters may also operate in a generator state. If these drive frequency converters are interconnected via a DC bus, the regenerative energy generated by one or more motors can be absorbed by other motors in the form of electrical energy. See the diagram below. As shown in the diagram, M1 and M3 are in a generator state, while M2 and M4 are in an electric motor state, absorbing energy. This is a highly efficient operating mode; even if multiple motors are continuously generating energy, there is no need to consider the method of absorbing regenerative energy. However, if rapid braking or emergency stopping is required, a braking unit and braking resistor must be added to release excess energy during dynamic transitions. Analysis of the operation process of two rigidly connected motors: In many cases, two motors will pull the same object. For example, a crane has two motors pulling the crane from both ends. See the diagram below. For the control of such a two-motor speed control system, there are generally two forms: one frequency converter pulling two motors; two frequency converters each pulling one motor. What are the characteristics of these two methods? For the first method, see the wiring diagram below. The advantages of this method are: the frequency settings of the two motors are exactly the same, speed adjustment is convenient and easy, and acceleration and deceleration are completely synchronized; the disadvantages are: the speed error caused by the difference in characteristics between the two motors cannot be corrected, which will result in different outputs of the two motors, causing one motor to be under light load while the other is under overload, eventually causing the overloaded motor to trip and the system to collapse. Therefore, if this system speed regulation is to be used, the characteristics of the two motors must be basically the same. For the second method, the basic wiring diagram is shown below: When working in this way, as long as the parameters of the two frequency converters are adjusted, the above shortcomings can be made up for. Let's first look at the impact of the different characteristics of the two motors on the output torque. As shown in the figure below, if the settings of the two frequency converters are the same, the output frequency is basically the same (ignoring the difference caused by the analog-to-digital conversion of the frequency converter). Since the motors are rigidly connected, the operating speed is the same. As can be seen from the figure, the difference in characteristics causes the difference in output torque. As shown in the figure above, at the same speed ne, the output torque T2 of motor M2 is greater than the output torque T1 of motor M1. For situations like inconsistent output, an effective solution is to increase the setpoint frequency of M1 (with a soft output characteristic), as shown by the dotted line in the diagram above. This allows M1 to output the same torque as M2 at speed ne, ensuring uniform load distribution. However, in practical applications, we cannot directly obtain the motor's characteristic curve. Therefore, in actual applications, the frequency is typically adjusted by measuring the current. This involves increasing the output frequency of the inverter with the lower output current, continuously observing the output current of both inverters until they are consistent. Finally, this adjustment is set in parameter b7-01 of the inverter and added to the setpoint. Note: b7-01: Slip amount at the highest output frequency when rated torque occurs (softening the characteristics). The adjustment range of the DROOP gain for parameter b7-01 of the Yolico inverter is 0 to +100%. In the rigid connection system mentioned above, we can also give a certain DROOP gain to the motor with stiff characteristics. This way, when the motor load is high, the output speed will automatically decrease to accommodate the motor with soft characteristics, automatically adjusting the load balance. In fact, after giving a certain DROOP gain compensation, the stiffness of the motor is artificially reduced, making the characteristics of the two motors more consistent. The DC bus sharing experiment was conducted to verify that when the motor has regenerative energy, the DC bus voltage of the inverter will increase, while when the DC bus is shared, the DC bus voltage will be suppressed. A system as shown in the figure below was constructed. Since the system uses a synchronous belt and synchronous pulley, there is no slip between them, so the system is similar to a rigid system. The conclusions drawn from this experiment are also applicable to rigid systems. I. DC Voltage When the DC Bus of the Inverter is Not Shared: 1. When stationary, the DC bus voltages of the two inverters are 531V and 531V respectively; under no-load conditions, when inverter M1 operates at 12Hz and inverter M2 is stopped, the DC bus voltages are 525V and 531V respectively; when inverter M2 operates at 12Hz and inverter M1 is stopped, the DC bus voltages are 531V and 525V respectively; as shown in the table below: When M1 and M2 are both stationary, M1 operates at 12Hz, M2 is stopped, M2 operates at 12Hz, M1 is stopped, M1 bus voltage Udc=531V Udc=525V Udc=531V M2 bus voltage Udc=531V Udc=531V Udc=525V From the above data, it is clear that the DC voltage will decrease slightly after the inverter is running under no-load conditions. 2. When both frequency converters are running at a frequency of 10Hz, the DC bus voltage is 525V. At this time, keeping the frequency of M2 constant, and gradually increasing the frequency of M1, the following set of data (unit: V) is obtained. +1Hz +2Hz +2.1Hz +2.2Hz +2.3Hz +2.4Hz +2.5Hz +2.6Hz M1 525 522 522 522 522 522 522 M2 525 549 550 552 571 592 609 627 Tension 2.4 3.3 +2.7Hz +2.8Hz +3Hz +3.1Hz 522 522 522 522 660 700 772 800 5.76 As shown in the table above, as the given frequency of M1 gradually increases, M2, being dragged along by M1, reaches a speed exceeding its own synchronous speed, generating regenerative energy. Since no braking resistor is connected, the DC bus voltage gradually increases, and the detected tension also gradually increases. When the voltage rises to 800V, the inverter trips due to overvoltage, and the system is protected and stops operating. Furthermore, we can conclude that the increased load did not cause the DC voltage of M1 to continue to drop. In other words, under rated load, the DC voltage only decreases slightly compared to the no-load condition and remains constant. Additionally, when the system is running smoothly, the counter-torque generated by M2 is equal in magnitude and opposite in direction to the torque of M1, as shown in the figure below. We summarize the DC voltage variation pattern when the DC bus is not shared as follows: If Fm1 > Fm2: The DC voltage of inverter M1 will decrease. (Fm: given frequency) The DC voltage of inverter M2 will increase. As shown in the figure below. II. DC Voltage When the DC Bus of Two Inverters is Shared After sharing the DC bus of two inverters, the above experiment was repeated. Inverter M2 was still given a frequency of 12Hz, and then the frequency of M1 was gradually increased. The following set of data was obtained: +0Hz +1Hz +2Hz +3Hz +4Hz +5Hz +6Hz M1 528 526 527 526 526 523 526 M2 528 526 527 526 526 523 526 Tension 0 3.9 6.3 7.7 8.8 9.0 As can be seen from the table above, after sharing the DC bus, the DC voltage of M2 no longer increases. As the frequency of M1 increases, the tension of the synchronous belt gradually increases, and the load on the motor also gradually increases, resulting in stable system operation. From the above experiment, it can be concluded that sharing the DC bus effectively utilizes the regenerative energy of the motor, suppresses the DC bus voltage, and improves the efficiency of the entire system. In systems with multiple motors driving the same load, this is an effective solution, and it is also frequently used in systems with several motors operating synchronously. Let's look at the tension data. The experiment shows that when the frequency of M1 increases by 3Hz, the tension voltage is 5.76V when the bus is not shared, while it is 6.3V when the bus is shared. The reason for this difference is that when the DC voltage increases, the rectifier bridge will turn off and cannot absorb energy from the power supply, resulting in a smaller braking torque than when the bus is shared. On the other hand, when the DC voltage increases, the charging current of the intermediate capacitor generated by the motor will decrease, resulting in a smaller braking torque than when the bus is shared. Therefore, the tension is greater when the bus is shared than when it is not. Points to note and connection conditions for DC bus sharing: Not all frequency converters can be unconditionally connected to a shared bus. The following instructions are for Yolico frequency converters. For other brands of frequency converters, please refer to the relevant manuals. 1. Specify the requirement when ordering: DC bus sharing. 2. Capacity ≤ 37kW: Direct parallel connection is allowed. 3. Consider the charging time constant of the controllable rectification. Inverters with a power of 45kW and above use controllable rectification. In this case, consider the DC shared connection time. Generally, the inverter with the longest charging time constant should be used as the standard, and both should be connected to the DC shared bus simultaneously to reduce the impact of DC current and large circulating current. 4. Avoid connecting more than two inverters with different power ratings to the DC bus as much as possible to prevent the rectifier section of the smaller inverter from charging the larger inverter and burning out the rectifier section of the smaller inverter. 5. If possible, use the following wiring method. This is a wiring method that conforms to industry standards. The function of the DC reactor: to suppress di/dt and protect the rectifier module. If the inverter power is small and the capacity is basically the same, and they are powered on at the same time, the DC reactor can be omitted. The function of the fast-acting fuse: to protect the rectifier module from damage caused by a short circuit at the DC terminal. Generally, this is only caused by a conductive object falling into the inverter, so it can be omitted in most cases. 6. When using distributed power supply, ensure that all equipment is powered on simultaneously. 7. During system operation, ensure that the bus current does not exceed the rated current corresponding to the capacity of each piece of equipment.