论文部分内容阅读
普荷夫曾由建筑力学中的位移法的系数公式(?)(?)_1=l/(3EJ)M_(12)+l/(6EJ)M_(21)+Ψ+(ωv)/(EJ)(?)_2=-l/(6EJ)M_(12)-l/(3EJ)M_(21)+Ψ-(ωu)/(EJ)求得受弯杆件(图1)的T 形代值电路(图3),并以解电路的各种方法去解决相应的力学问题,很有成效。普荷夫还指出了用形代值电路及用实验方法研究的可能性。依照普夫的指示,本文首先求出受弯杆件的形代值电路(图5),并比较T 形代值电路与形代值电路的用途。结果证明:形代值电路对力学中的初参数法特别合用,对力学中的其他方法则以用T 形代值电路为宜。本文扩大了参数表,包括更多类型的荷重,因此扩大了代值电路法的应用范围,例如计算温度变化的影响等。在比较各种电路解法和力学解法的基础上,发现所有的力学方法都有一电路解法与之对应,而且电路解法还能突破相应的力学方法所受的限制。本文的主要部分是实验研究方面。由于T 形代值电路和形代值电路都含有负电阻(静态的),无法用来进行实验研究。本文首先根据四端网络的原理找出T 形代值电路的不含负电阻的两种等效电路:X 形电路(图10)和桥T 形电路(图13)。比较这两种能够用来进行实验研究的代值电路所用的元件的价格,表明X 形代值电路宜用于作特殊电气模型,桥T 形代值电路宜用于作万用电气模型——计算台。进行实验研究时,曾注意:1)电源内阻的影响;2)参数的直线性;3)参数大小的影响;4)测量仪器的影响。实验结果和理论计算都表明:X 形代值电路中的误差全都是由于测量仪器所引起的,元件本身和电源内阻的影响是极微的。测量电压要用补偿法才能保证较高的准确度,安培计的影响可以用加大元件的电阻数值的办法去削弱。至于桥T 形代值电路中的误差则主要是由于所含理想变压器的条件不能满足所引起的,在工频中实难克服。因此,采用X 形代值电路作为电气模型的基础。用简单杆件的电气模型进行实验的结果,证明误差在0.5%以下,用复杂杆件系统的电气模型进行实验的结果,证明误差小于1%。本文最后一部分是关于解决10根杆件以下的任何杆件系统的弯曲问题的计算台的设计。
Pheff was composed of the coefficient formula (?) (?) _ 1 = l / (3EJ) M_ (12) + l / (6EJ) M_ (21) + Ψ + (ωv) / The T-generation of the bent rod (Fig. 1) is obtained by () _ _ 2 = - 1 / (6EJ) M_ (12) -l / (3EJ) M_ (21) + Ψ- (ωu) Value circuit (Figure 3), and to solve the circuit in various ways to solve the corresponding mechanical problems, very effective. Phevl also pointed out the possibility of using algebraic circuits and experimental methods. In accordance with Pf’s instructions, this paper first find the value of the rod-shaped parts of the value of the circuit (Figure 5), and compare the T-value generation circuits and generational value circuits. The result proves that the algebraic value circuit is especially suitable for the first parameter method in mechanics and the T-value algebraic value circuit for the other methods in mechanics. This article expands the list of parameters, including more types of loads, thus broadening the range of applications of the algebraic circuits, such as calculating the effects of temperature changes. Based on the comparison of various circuit solutions and mechanics solutions, it is found that all the mechanical methods have a circuit solution corresponding to them, and the circuit solution can also break through the limitations of the corresponding mechanical methods. The main part of this article is about experimental research. Since both T-value generation circuits and generation value circuits contain negative resistance (static), they can not be used for experimental research. In this paper, we firstly find out two equivalent circuits of T-value generation circuits without negative resistance according to the principle of four-terminal network: X-shaped circuit (Figure 10) and bridge T-shaped circuit (Figure 13). Comparing the prices of the components used in these two kinds of algebraic circuits that can be used for experimental research shows that the X-value generation circuit should be used as a special electrical model and the bridge T-value generation circuit should be used as a universal electrical model - Computing platform. When conducting experimental studies, we have noticed: 1) the influence of the internal resistance of the power supply; 2) the linearity of the parameters; 3) the influence of the parameter size; and 4) the influence of the measuring instrument. The experimental results and theoretical calculations show that the errors in the X-value generation circuit are all caused by the measuring instruments. The influence of the components themselves and the internal resistance of the power supply is minimal. Measurement of voltage compensation method used to ensure high accuracy, the impact of the ammeter can increase the value of the component resistance to weaken. As for the bridge T-generation value of the error in the circuit is mainly due to the conditions contained in the ideal transformer can not meet caused by the power frequency is hard to overcome. Therefore, the X-value generation circuit is used as the basis of the electrical model. The experimental results of the electrical model with a simple rod show that the error is below 0.5%. The experimental results using an electrical model of a complex rod system show that the error is less than 1%. The last part of this article is about the design of a computer table that solves the bending problem of any bar system below 10 bars.